U.S. patent application number 10/883233 was filed with the patent office on 2005-03-31 for electrophysiological atlas and applications of same.
This patent application is currently assigned to Vanderbilt University. Invention is credited to Cetinkaya, Ebru, D'Haese, Pierre-Francois Dominique, Dawant, Benoit M., Kao, Changquing C., Konrad, Peter E..
Application Number | 20050070781 10/883233 |
Document ID | / |
Family ID | 33418352 |
Filed Date | 2005-03-31 |
United States Patent
Application |
20050070781 |
Kind Code |
A1 |
Dawant, Benoit M. ; et
al. |
March 31, 2005 |
Electrophysiological atlas and applications of same
Abstract
A method of creating an atlas that contains electrophysiological
information related to at least one of a plurality of living
subjects. In one embodiment, the method includes the steps of
choosing a brain image volume as a common image volume of
reference, acquiring electrophysiological information for a target
of interest, relating the acquired electrophysiological information
to spatial coordinates in the brain image volume of the target of
interest, and registering the brain image volume of the target of
interest to the common image volume of reference so as to create an
atlas in which any spatial coordinates of the brain of the target
of interest are related to atlas coordinates in the atlas such that
the acquired electrophysiological information associated with the
related spatial coordinates in the brain image volume of the target
of interest can be related to atlas coordinates in the atlas, and
vice versa.
Inventors: |
Dawant, Benoit M.;
(Nashville, TN) ; D'Haese, Pierre-Francois Dominique;
(Nashville, TN) ; Konrad, Peter E.; (Old Hickory,
TN) ; Kao, Changquing C.; (Brentwood, TN) ;
Cetinkaya, Ebru; (Eskischir, TR) |
Correspondence
Address: |
MORRIS MANNING & MARTIN LLP
1600 ATLANTA FINANCIAL CENTER
3343 PEACHTREE ROAD, NE
ATLANTA
GA
30326-1044
US
|
Assignee: |
Vanderbilt University
Nashville
TN
|
Family ID: |
33418352 |
Appl. No.: |
10/883233 |
Filed: |
July 1, 2004 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10883233 |
Jul 1, 2004 |
|
|
|
10833504 |
Apr 28, 2004 |
|
|
|
60466219 |
Apr 28, 2003 |
|
|
|
Current U.S.
Class: |
600/407 ;
607/45 |
Current CPC
Class: |
A61B 34/10 20160201;
A61B 90/10 20160201; A61N 1/36067 20130101; A61N 1/3605
20130101 |
Class at
Publication: |
600/407 ;
607/045 |
International
Class: |
A61B 005/05 |
Claims
What is claimed is:
1. A system for creating an atlas for optimal placement of a deep
brain stimulator in a brain of a target of interest, comprising: a.
a data storage device; b. an image acquisition device for acquiring
a brain image volume from the brain of the target of interest; c. a
data acquisition device for acquiring electrophysiological
information from the brain of the target of interest; and d. a data
processing device operably coupled to the data storage device, the
image acquisition device and the data acquisition device,
respectively, and performing the steps of: (a). relating the
acquired electrophysiological information to spatial coordinates in
the acquired brain image volume of the target of interest; and (b).
registering the acquired brain image volume of the target of
interest to a common image volume of reference so as to create an
atlas in which spatial coordinates of the brain of the target of
interest are related to atlas coordinates such that the acquired
electrophysiological information associated with the related
spatial coordinates in the acquired brain image volume of the
target of interest can be related to atlas coordinates in the
atlas, and vice versa.
2. The system of claim 1, wherein the data processing device
further performs the step of storing the atlas in a digitized
format in the data storage device.
3. The system of claim 1, wherein the atlas has an architecture to
be accessible over a network.
4. The system of claim 3, further comprising a user interface in
communication with the atlas.
5. The system of claim 4, wherein the user interface is used for
populating the atlas with new electrophysiological information
acquired from a target of interest.
6. The system of claim 4, wherein the user interface is used for
accessing the electrophysiological information from the atlas.
7. The system of claim 6, wherein the user interface is used for
obtaining the electrophysiological information from the atlas in
one of a text format, an image format and a mixture thereof.
8. The system of claim 7, wherein the obtained electrophysiological
information includes an initial optimal position for at least one
deep brain stimulator to be, placed in a brain of a target of
interest.
9. The system of claim 1, wherein the data storage device comprises
a memory.
10. The system of claim 1, wherein the image acquisition device is
arranged, in use, to acquire a computerized tomographical image
and/or a magnetic resonance image for a target of interest.
11. The system of claim 1, wherein the data acquisition device
comprises at least one microelectrode placed in a brain of a target
of interest.
12. The system of claim 1, wherein the data acquisition device
comprises at least one stimulation electrode placed in a brain of a
target of interest.
13. The system of claim 1, wherein the data acquisition device
comprises at least one deep brain stimulator placed in a brain of a
target of interest.
14. The system of claim 1, wherein the electrophysiological
information comprises pre-operative information, intra-operative
information and post-operative information for a target of
interest, respectively.
15. The system of claim 14, wherein the pre-operative information
comprises at least one piece of information associated with
presenting complaints, locations of symptoms related to one or more
diseases, type and degree of the one or more diseases, unified
Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof.
16. The system of claim 15, wherein the symptoms comprise at least
one of upper extremity rigidity, lower extremity rigidity, upper
extremity dystonia and lower extremity dystonia.
17. The system of claim 15, wherein the unified Parkinson's disease
rating scale scores are a rating tool for evaluating mentation,
behavior and mood, activities of daily living, motor activity, and
complication of therapy for a target of interest undergoing
treatment.
18. The system of claim 14, wherein the intra-operative information
comprises at least one piece of information associated with at
least one microelectrode, wherein the information includes
microelectrode recordings, a position of the microelectrode
recordings, a label of a structure in which the microelectrode
recordings is located, and any mixture thereof.
19. The system of claim 18, wherein the microelectrode recordings
are characterized by a firing rate that measures tonic activity and
indices that measures phasic activity, wherein the indices include
a burst index, a pause ratio, a pause index, and an interspike
interval histogram.
20. The system of claim 14, wherein the intra-operative information
comprises at least one piece of information associated with at
least one stimulation electrode, wherein the information includes
voltages applied to the at least one stimulation electrode, a
response of a target of interest undergoing treatment to the
stimulation, differences in voltage between disappearance of
symptoms and appearance of side effects, a position of the at least
one stimulation electrode, a final intra-operative target position
of a deep brain stimulator to be placed, and any mixture
thereof.
21. The system of claim 20, wherein the response of the target of
interest undergoing treatment to the stimulation includes loss of
rigidity, location where the loss of rigidity is observed,
appearance of side effects, and/or location affected by these side
effects.
22. The system of claim 14, wherein the post-operative information
comprises at least one piece of information associated with at
least one deep brain stimulator, wherein the information includes a
position of the at least one deep brain stimulator in
post-operative computerized tomographical images, optimal setting
of the at least one deep brain stimulator, overall assessment of a
target of interest after placement of the at least one deep brain
stimulator, and any mixture thereof.
23. A method of creating an atlas that contains
electrophysiological information related to at least one of a
plurality of living subjects, wherein any portion of a brain of one
of the plurality of living subjects and corresponding brain image
volume may be identified by a set of corresponding spatial
coordinates, comprising the steps of: a. choosing a brain image
volume as a common image volume of reference from a plurality of
brain image volumes, each of the plurality of brain image volumes
being acquired pre-operatively from a brain of one of the plurality
of the living subjects; b. acquiring electrophysiological
information for one of the plurality of living subjects; c.
relating the acquired electrophysiological information to spatial
coordinates in the brain image volume of the corresponding living
subject; and d. registering the brain image volume of the
corresponding living subject to the common image volume of
reference so as to create an atlas in which any spatial coordinates
of the brain of the corresponding living subject are related to
atlas coordinates in the atlas such that the acquired
electrophysiological information associated with the related
spatial coordinates in the brain image volume of the corresponding
living subject can be related to atlas coordinates in the atlas,
and vice versa.
24. The method of claim 23, further comprising the step of storing
the atlas in a digitized format.
25. The method of claim 23, wherein the atlas has an architecture
to be accessible over a network.
26. The method of claim 25, further comprising the step of
populating the atlas with new electrophysiological information
acquired from a target of interest.
27. The method of claim 25, further comprising the step of
accessing the electrophysiological information from the atlas.
28. The method of claim 27, further comprising the step of
obtaining the electrophysiological information from the atlas in
one of a text format, an image format and a mixture thereof.
29. The method of claim 28, wherein the obtained
electrophysiological information includes an initial optimal target
position for at least one deep brain stimulator to be placed in a
brain of a target of interest.
30. The method of claim 23, wherein the electrophysiological
information comprises pre-operative information, intra-operative
information and post-operative information for each of the
plurality of living subjects, respectively.
31. The method of claim 30, wherein the pre-operative information
comprises at least one piece of information associated with
presenting complaints, locations of symptoms related to one or more
diseases, type and degree of the one or more diseases, unified
Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof.
32. The method of claim 31, wherein the symptoms comprise at least
one of upper extremity rigidity, lower extremity rigidity, upper
extremity dystonia and lower extremity dystonia.
33. The method of claim 31, wherein the unified Parkinson's disease
rating scale scores are a rating tool for evaluating mentation,
behavior and mood, activities of daily living, motor activity, and
complication of therapy for a target of interest undergoing
treatment.
34. The method of claim 30, wherein the intra-operative information
comprises at least one piece of information associated with at
least one microelectrode, wherein the information includes
microelectrode recordings, a position of the microelectrode
recordings, a label of a structure in which the microelectrode
recordings is located, and any mixture thereof.
35. The method of claim 34, wherein the microelectrode recordings
are characterized by a firing rate that measures tonic activity and
indices that measures phasic activity, wherein the indices include
a burst index, a pause ratio, a pause index, and an interspike
interval histogram.
36. The method of claim 30, wherein the intra-operative information
comprises at least one piece of information associated with at
least one stimulation electrode, wherein the information includes
voltages applied to the at least one stimulation electrode, a
response of a target of interest undergoing treatment to the
stimulation, differences in voltage between disappearance of
symptoms and appearance of side effects, a position of the at least
one stimulation electrode, a final intra-operative target position
of a deep brain stimulator to be placed, and any mixture
thereof.
37. The method of claim 36, wherein the response of the target of
interest undergoing treatment to the stimulation includes loss of
rigidity, location where the loss of rigidity is observed,
appearance of side effects, and/or location affected by these side
effects.
38. The method of claim 30, wherein the post-operative information
comprises at least one piece of information associated with at
least one deep brain stimulator, wherein the information includes a
position of the at least one deep brain stimulator in
post-operative computerized tomographical images, optimal setting
of the at least one deep brain stimulator, overall assessment of a
target of interest after placement of the at least one deep brain
stimulator, and any mixture thereof.
39. A system that contains electrophysiological information related
to at least one of a plurality of living subjects, wherein any
portion of interest in the brain of one of the plurality of living
subjects and corresponding brain image volume may be identified by
a set of corresponding spatial coordinates, comprising: a. a data
storage device; and b. an atlas stored in the data storage device
for containing the electrophysiological information, the atlas
being created such that when a brain image volume is registered to
the atlas, any spatial coordinates of the brain image volume are
related to corresponding atlas coordinates in the atlas, and vice
versa.
40. The system of claim 39, wherein the atlas has an architecture
to be accessible over a network.
41. The system of claim 40, wherein the atlas can be in
communication with a user interface.
42. The system of claim 41, wherein the user interface is used for
populating the system with new electrophysiological information
acquired from a target of interest.
43. The system of claim 41, wherein the user interface is used for
accessing the electrophysiological information from the atlas.
44. The system of claim 39, wherein the electrophysiological
information comprises pre-operative information, intra-operative
information and post-operative information for each of the
plurality of living subjects, respectively.
45. The system of claim 44, wherein the pre-operative information
comprises at least one piece of information associated with
presenting complaints, locations of symptoms related to one or more
diseases, type and degree of the one or more diseases, unified
Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof.
46. The system of claim 45, wherein the symptoms comprise at least
one of upper extremity rigidity, lower extremity rigidity, upper
extremity dystonia and lower extremity dystonia.
47. The system of claim 46, wherein the unified Parkinson's disease
rating scale scores are a rating tool for evaluating mentation,
behavior and mood, activities of daily living, motor activity, and
complication of therapy for a target of interest undergoing
treatment.
48. The system of claim 44, wherein the intra-operative information
comprises at least one piece of information associated with at
least one microelectrode, wherein the information includes
microelectrode recordings, a position of the microelectrode
recordings, a label of a structure in which the microelectrode
recordings is located, and any mixture thereof.
49. The system of claim 48, wherein the microelectrode recordings
are characterized by a firing rate that measures tonic activity and
indices that measures phasic activity, wherein the indices include
a burst index, a pause ratio, a pause index, and an interspike
interval histogram.
50. The system of claim 44, wherein the intra-operative information
comprises at least one piece of information associated with at
least one stimulation electrode, wherein the information includes
voltages applied to the at least one stimulation electrode, a
response of a target of interest undergoing treatment to the
stimulation, differences in voltage between disappearance of
symptoms and appearance of side effects, a position of the at least
one stimulation electrode, a final intra-operative target position
of a deep brain stimulator to be placed, and any mixture
thereof.
51. The system of claim 50, wherein the response of the target of
interest undergoing treatment to the stimulation includes loss of
rigidity, location where the loss of rigidity is observed,
appearance of side effects, and/or location affected by these side
effects.
52. The system of claim 44, wherein the post-operative information
comprises at least one piece of information associated with at
least one deep brain stimulator, wherein the information includes a
position of the at least one deep brain stimulator in
post-operative computerized tomographical images, optimal setting
of the at least one deep brain stimulator, overall assessment of a
target of interest after placement of the at least one deep brain
stimulator, and any mixture thereof.
53. The system of claim 39, wherein the data storage device
comprises a memory.
54. The system of claim 39, further comprising a controller in
communication with the data storage device.
55. A computer readable medium or media, comprising: a. a data
structure relating to an atlas that contains electrophysiological
information related to at least one of a plurality of living
subjects, wherein any portion of interest for the brain of at least
one of the plurality of living subjects and corresponding image
volume may be identified by a set of corresponding spatial
coordinates; and b. a user interface in communication with the data
structure.
56. The computer readable medium or media of claim 55, wherein the
data structure comprises a plurality of transformations and
corresponding inverses of the plurality of transformations.
57. The computer readable medium or media of claim 56, wherein each
of a plurality of transformations registers a brain image volume to
the atlas, wherein the brain image volume is acquired from one of
the plurality of living subjects.
58. The computer readable medium or media of claim 57, wherein the
atlas is created such that when a brain image volume of a living
subject is registered to the atlas, any spatial coordinates in the
brain image volume of the living subject are related to
corresponding atlas coordinates in the atlas, and vice versa.
59. The computer readable medium or media of claim 58, wherein the
electrophysiological information associated with spatial
coordinates from which the electrophysiological information is
acquired in the brain of the living subject can be related to atlas
coordinates in the atlas, and vice versa.
60. The computer readable medium or media of claim 59, wherein the
user interface is used for populating the atlas with new
electrophysiological information acquired from a target of
interest.
61. The computer readable medium or media of claim 59, wherein the
user interface is used for accessing the electrophysiological
information from the atlas.
62. The computer readable medium or media of claim 61, wherein the
user interface is used for obtaining the electrophysiological
information from in the atlas in one of a text format, an image
format and a mixture thereof.
63. The computer readable medium or media of claim 62, wherein the
obtained electrophysiological information includes an initial
optimal target position for at least one deep brain stimulator to
be placed in a brain of a target of interest.
64. The computer readable medium or media of claim 57, wherein the
atlas is stored in a digitized format of files.
65. The computer readable medium or media of claim 57, wherein the
electrophysiological information comprises pre-operative
information, intra-operative information and post-operative
information for each of the plurality of living subjects,
respectively.
66. A method for optimal placement of a deep brain stimulator in a
brain of a target of interest, comprising the steps of: a.
acquiring from the target of interest pre-operatively at least one
piece of information associated with a state of brain condition of
the target of interest; b. accessing remotely an atlas that
contains electrophysiological information related to a plurality of
living subjects, wherein any portion of the brain of one of the
plurality of living subjects and corresponding brain image volume
may be identified by a set of corresponding spatial coordinates; c.
entering the acquired information from the target of interest to
the atlas to find a match between the acquired information and the
electrophysiological information contained in the atlas; and d.
automatically obtaining an optimal position in the brain of the
target of interest for placing a deep brain stimulator from the
matched information.
67. The method of claim 66, wherein the atlas is stored in a memory
device associated with a central host computer.
68. The method of claim 67, wherein the accessing step comprises
the step of accessing the atlas over a network from a client
computer, and wherein the central host computer and client computer
are coupled to and in communication with the network,
respectively.
69. The method of claim 68, wherein the network comprises at least
one of a public network, a dedicated network, a local network, and
any combination of them.
70. The method of claim 69, wherein the public network comprises
the Internet.
71. The method of claim 66, further comprising the steps of
acquiring electrophysiological information intra-operatively from
the brain of the target of interest and adjusting the optimal
position in the brain of the target of interest for placing a deep
brain stimulator accordingly.
72. The method of claim 71, further comprising the steps of finding
a final optimal position in the brain of the target of interest for
placing a deep brain stimulator from the adjusted optimal position
and placing a deep brain stimulator in the brain of the target of
interest.
73. The method of claim 7 1, further comprising the steps of
downloading information related to the optimal position in the
brain of the target of interest for placing a deep brain stimulator
from the matched information to a local computer and adjusting the
optimal position in the brain of the target of interest for placing
a deep brain stimulator from the downloaded information.
74. The method of claim 66, further comprising the step of updating
the atlas from information related to the target of interest.
75. The method of claim 66, wherein the information associated with
the state of brain condition of the target of interest comprises
presenting complaints, locations of symptoms related to one or more
diseases, type and degree of the one or more diseases, unified
Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof of the target of
interest.
76. The method of claim 66, wherein the state of brain condition is
related to a type of a disease.
77. The method of claim 76, wherein the state of brain condition is
related to a degree of a disease.
78. A method for optimal placement of a deep brain stimulator in a
brain of a target of interest, comprising the steps of: a.
acquiring from the target of interest pre-operatively at least one
piece of information associated with a state of brain condition of
the target of interest; b. accessing remotely an atlas that
contains electrophysiological information related to a plurality of
living subjects, wherein the atlas is formed with a plurality of
clusters, each cluster being related to a state of brain condition
and having a plurality of optimal positions for a deep brain
stimulator distributed therein; c. entering the acquired
information from the target of interest to the atlas to find a
match between the acquired information and the electrophysiological
information contained in the atlas; and d. automatically obtaining
an optimal position in the brain of the target of interest for
placing a deep brain stimulator from one of the plurality of
optimal positions for a deep brain stimulator distributed in the
plurality of clusters.
79. The method of claim 78, wherein the state of brain condition is
related to a type of a disease.
80. The method of claim 79, wherein the state of brain condition is
related to a degree of a disease.
81. The method of claim 78, wherein the information associated with
the state of brain condition of the target of interest comprises
presenting complaints, locations of symptoms related to one or more
diseases, type and degree of the one or more diseases, unified
Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof of the target of
interest.
Description
CROSS-REFERENCE TO RELATED PATENT APPLICATION
[0001] This application is a continuation-in-part of U.S. patent
application Ser. No. 10/833,504, filed Apr. 28, 2004, entitled
"APPARATUS AND METHODS OF OPTIMAL PLACEMENT OF DEEP BRAIN
STIMULATOR," by Benoit M. Dawant, the disclosure for which is
incorporated herein by reference in its entirety, which itself
claims the benefit, pursuant to 35 U.S.C. .sctn. 119(e), of
provisional U.S. patent application Serial No. 60/466,219, filed
Apr. 28, 2003, entitled "APPARATUS AND METHODS OF COMPUTERIZED
ATLAS-GUIDED POSITIONING OF DEEP BRAIN STIMULATORS," by Benoit M.
Dawant, which is incorporated herein by reference in its
entirety.
[0002] Some references, which may include patents, patent
applications and various publications, are cited and discussed in
the description of this invention. The citation and/or discussion
of such references is provided merely to clarify the description of
the present invention and is not an admission that any such
reference is "prior art" to the invention described herein. All
references cited and discussed in this specification are
incorporated herein by reference in their entireties and to the
same extent as if each reference was individually incorporated by
reference. In terms of notation, hereinafter, "[n]" represents the
nth reference cited in the reference list. For example, [9]
represents the 9th reference cited in the reference list, namely,
G. Rhode, A. Aldroubi and B. M. Dawant, "The Adaptive-bases
algorithm for intensity-based nonrigid image registration," IEEE
Transactions on Medical Imaging, vol. 22, no. 11, pp 1470-1479,
2003.
FIELD OF THE INVENTION
[0003] The present invention generally relates to an atlas, and in
particular to the creation and/or utilization of an atlas that
contains electrophysiological information related to one or more
living subjects for optimal placement of one or more deep brain
stimulators placement in a brain of a target of interest.
BACKGROUND OF THE INVENTION
[0004] Since its first Food and Drug Administration (hereinafter
"FDA") approval in 1998, deep-brain stimulation (hereinafter "DBS")
has gained significant popularity in the treatment of a variety of
brain-controlled disorders, including movement disorders [1, 2].
The therapy of the DBS has significant applications in the
treatment of tremor, rigidity, and drug induced side effects in
patients with Parkinson's disease and essential tremor. Generally,
such treatment involves placement of a DBS electrode lead through a
burr hole drilled in the patient's skull, followed by placement of
the electrode lead and then applying appropriate stimulation
signals through the electrode lead to the physiological target. The
placement portion of the treatment, involving stereotactic
neurosurgical methodology, is very critical, and has been the
subject of much attention and research. In particular, finding the
deep brain target and then permanently placing the electrode lead
so that it efficiently stimulates such target is very
important.
[0005] Yet finding the optimal physiological target in deep brain
stimulation implants for the treatment of movement disorders is a
particularly complicated task. This is especially true for the
treatment of symptoms that cannot be tested at the operating table
during the electrode lead implantation. For instance, it is
practically impossible to test walking and postural stability in
Parkinson's Disease (hereinafter "PD") patients during the DBS lead
implantation. Two other major PD symptoms, Rigidity and Akinesia,
are also considered difficult to evaluate quantitatively during DBS
lead implantation. On the other hand, the surgical targets of
interest involve deep brain nuclei or subregions within the
subthalamus or globus pallidus intemus. These structures are not
visible in any current imaging modalities, such as magnetic
resonance imaging (hereinafter "MRI"), X-ray computerized
tomography (hereinafter "CT"), or Positron Emission Tomography
(hereinafter "PET").
[0006] Ideally, the optimal target for the DBS therapy should be
located within the stimulation range of 1 or 2 contacts, each
contact measuring 1.5 mm separated by either 1.5 mm or 0.5 mm.
Effective stimulation results when the contacts surround the
optimal target [3, 4]. For example, as shown in FIG. 1, for
placement of a 4-contact electrode lead of a deep brain stimulator
100, which has a tip portion 170, a central body portion 150 and
associated contacts 110, 120, 130 and 140 (Medtronic #3387 or #3389
quadripolar lead.RTM., Medtronic, Inc., Minneapolis, Minn.), in the
proximity of functional areas which one may refer to as targets or
targeted regions, a preferable scenario is that two contacts 110
and 120 of the quadripolar lead 100 lie above and the other
contacts 130 and 140 lie below a target. For this example of the
lead, each of the contacts 110, 120, 130, and 140 has a length,
d.sub.1, which is substantially around 1.5 mm for a Medtronic #3387
or #3389 quadripolar lead.RTM., and the distance between two
neighboring contacts, for example, 130 and 140, is d.sub.2, where
d.sub.2=1.5 mm for Medtronic #3387 quadripolar lead.RTM., and
d.sub.2=0.5 mm for Medtronic #3389 quadripolar lead.RTM.,
respectively. If the contacts are located as little as 2 mm away
from the desired target, ineffective stimulation results, which may
be due to several reasons: (i) failure to capture control of the
group of neurons, (ii) stimulation of non-desirable areas resulting
in unpleasant stimulation, or (iii) necessity for higher stimulus
intensities to produce the desired effect resulting in reduced
battery life of the implantation, or an any combination of these or
other reasons. At least for these reasons, targeting the specific
neurons of interest for the DBS therapy requires millimetric
precision and allowance for variability among patients. Therefore,
the process of implantation of a DBS electrode lead requires
stereotactic neurosurgical methodology, i.e., the use of a common
reference coordinate system to target structures within the brain
of a target of interest. Typically, the process of implantation of
a DBS electrode follows a step-wise progression of (i) initial
estimation of target localization based on imaged anatomical
landmarks, (ii) intra-operative microanatomical mapping of key
features associated with the intended target or target position of
the brain of a target of interest, (iii) adjustment of the final
target of implantation by appropriate shifts in three dimensional
space, and (iv) implantation of a quadripolar electrode with
contacts located surrounding the final desired target or target
position of the brain of the target of interest.
[0007] Because of the invisibility of deep brain targets or target
positions of interest in any current imaging modalities, such as
MRI, CT, or PET, the location of these targets can only be inferred
approximately from the position of adjacent structures that are
visible in the images. To augment the information that these images
provide, printed anatomic atlases or electronic versions of these
have been used. Anatomic atlases, such as the Schaltenbrand-Wahren
atlas [14], involve a series of unevenly spaced brain sections that
have been histologically stained to reveal the structures and
substructures of interest. When digitized, these atlases can be
superimposed on the pre-operative images using landmarks visible
both in the atlas and in the image volumes. Although it represents
a partial solution to the target identification problem, this
approach seems to suffer from a number of shortcomings [15]. First,
available anatomic atlases have been created from one single brain
[16] or from several hemispheres pertaining to different
individuals [14]. When a single brain is used, information is
limited to one sectioning plane per hemisphere. When several brains
are used, these atlases show non-contiguous anatomy in intersecting
orthogonal slices. Registration (i.e. spatial alignment) of these
atlases to the image volumes also raises a number of issues. The
standard procedure is to register atlas and image volumes using the
inter commisural anterior commissure (hereinafter "AC")-posterior
commissure (hereinafter "PC") reference system. This method is one
in which the AC and PC points are manually selected in the image
volumes. The image volumes are first translated to align the AC
points. They are then rotated to align the AC-PC line and the
midsagittal planes. Unfortunately, this technique results in
substantial misregistration errors. A better approach proposed by
St-Jean et al. [17] involves digitizing the Schaltenbrand-Wahren
atlas, stacking individual slices, and creating three-dimensional
(hereinafter "3D") structures from these slices through
interpolation. These 3D structures are then registered to one
magnetic resonance (hereinafter "MR") image volume by identifying
homologous landmarks, thus creating an MR image volume on which
labels from the atlas can be projected. But, this procedure only
guarantees that the landmarks are registered to each other. In a
later publication [15], the authors acknowledge that this
limitation plus the fact that the creation of the 3D structures
involves interpolating two-dimensional (hereinafter "2D") atlas
slices that can be between 0.5 mm and 3 mm apart limit the accuracy
and therefore the clinical usefulness of this approach.
[0008] In current clinical practice, the initial target
localization is manually selected on MR images based on AC-PC
coordinates. The initial target localization is refined by
intra-operatively probing a surrounding region of the initial
target with a recording and/or a stimulating electrode. First a
recording electrode is placed into the initial target localization
to characterize neuronal firing patterns, which are in turn used to
infer locations of deep brain nuclei relevant to the targeted
region. A stimulating electrode is then placed into the inferred
location to elicit responses in an awake patient. Both of these
sources of information allow neurosurgeons, neurologists, and
neurophysiologists to establish functional borders and to mentally
reconstruct a somatotopic organization of the structures of
interest so as to identify a final target location at which a deep
brain stimulator is to be placed. It can be a lengthy process
(sometimes extending for hours in the awake patient) and it
requires expertise in neurosurgery, neurophysiology, and clinical
neurology [18, 19]. This combined expertise is available only at a
limited number of sites, which limits access to the procedure to
about 3000 patients per year despite the estimated 180,000 patients
per year who would benefit from it in the United States alone.
Great clinical relevance would be gained if electrophysiological
information could be captured from a large population, processed,
and represented in a way that would make it usable for guidance
purposes.
[0009] Therefore, a heretofore unaddressed need exists in the art
to address the aforementioned deficiencies and inadequacies.
SUMMARY OF THE INVENTION
[0010] In one aspect, the present invention relates to a system for
creating an atlas for optimal placement of a deep brain stimulator
in a brain of a target of interest. In one embodiment, the system
includes a data storage device, an image acquisition device for
acquiring a brain image volume from the brain of the target of
interest, and a data acquisition device for acquiring
electrophysiological information from the brain of the target of
interest.
[0011] Furthermore, the system includes a data processing device
operably coupled to the data storage device, the image acquisition
device and the data acquisition device, respectively. The data
processing device is adapted for, among other things, performing
the steps of relating the acquired electrophysiological information
to spatial coordinates in the acquired brain image volume of the
target of interest, and registering the acquired brain image volume
of the target of interest to a common image volume of reference so
as to create an atlas in which spatial coordinates of the brain of
the target of interest are related to atlas coordinates such that
the acquired electrophysiological information associated with the
related spatial coordinates in the acquired brain image volume of
the target of interest can be related to atlas coordinates in the
atlas, and vice versa. The data processing device is adapted for
further performing the step of storing the atlas in a digitized
format of files in the data storage device.
[0012] Moreover, the system includes a user interface in
communication with the atlas, which has an architecture to be
accessible over a network. The user interface is used for
populating the atlas with new electrophysiological information
acquired from a target of interest, accessing the
electrophysiological information from the atlas, and obtaining the
electrophysiological information from in the atlas in one of a text
format, an image format and a mixture thereof, respectively. In one
embodiment, the obtained electrophysiological information includes
an initial optimal target position for at least one deep brain
stimulator to be placed in a brain of a target of interest.
[0013] In one embodiment, the data storage device includes a
memory. The image acquisition device is arranged, in use, to
acquire a computerized tomographical image and/or a magnetic
resonance image for a target of interest. In one embodiment, the
data acquisition device includes at least one microelectrode,
and/or at least one stimulation electrode placed in a brain of a
target of interest for acquiring intra-operative information for
the target of interest. The data acquisition device further
includes at least one deep brain stimulator placed in the brain of
the target of interest for acquiring post-operative
information.
[0014] The electrophysiological information to be acquired includes
pre-operative information, intra-operative information and
post-operative information for a target of interest, respectively.
In one embodiment, the pre-operative information includes at least
one piece of information associated with presenting complaints,
locations of symptoms related to one or more diseases, type and
degree of the one or more diseases, unified Parkinson's disease
rating scale scores both on and off medications, mini-mental status
examination, medications and dosages, cognitive performance, gait
performance, pre-operative target positions, and any mixture
thereof. The symptoms have at least one of upper extremity
rigidity, lower extremity rigidity, upper extremity dystonia and
lower extremity dystonia. The unified Parkinson's disease rating
scale scores are a rating tool for evaluating mentation, behavior
and mood, activities of daily living, motor activity, and
complication of therapy for a target of interest undergoing
treatment.
[0015] The intra-operative information includes at least one piece
of information associated with at least one microelectrode, where
the information includes microelectrode recordings, a position of
the microelectrode recordings, a label of a structure in which the
microelectrode recordings is located, and any mixture thereof. The
microelectrode recordings are characterized by a firing rate that
measures tonic activity and indices that measures phasic activity,
where the indices include a burst index, a pause ratio, a pause
index, and an interspike interval histogram. Other characteristic
features may also be extracted from the microelectrode recordings
and can also be utilized. The intra-operative information may also
includes at least one piece of information associated with at least
one stimulation electrode, where the information includes voltages
applied to the at least one stimulation electrode, a response of a
target of interest undergoing treatment to the stimulation,
differences in voltage between disappearance of symptoms and
appearance of side effects, a position of the at least one
stimulation electrode, a final intra-operative target position of a
deep brain stimulator to be placed, and any mixture thereof. The
response of the target of interest undergoing treatment to the
stimulation includes loss of rigidity, location where the loss of
rigidity is observed, appearance of side effects, and/or location
affected by these side effects.
[0016] The post-operative information comprises at least one piece
of information associated with at least one deep brain stimulator,
where the information includes a position of the at least one deep
brain stimulator in post-operative computerized tomographical
images, optimal setting of the at least one deep brain stimulator,
overall assessment of a target of interest after placement of the
at least one deep brain stimulator, and any mixture thereof.
[0017] In another aspect, the present invention relates to a method
of creating an atlas. The atlas contains electrophysiological
information related to at least one of a plurality of living
subjects, where any portion of interest of a brain of one of the
plurality of living subjects and corresponding image volume may be
identified by a set of corresponding spatial coordinates.
[0018] In one embodiment, among other things, the method includes
the step of choosing a brain image volume as a common image volume
of reference from a plurality of brain image volumes, each of the
plurality of brain image volumes being acquired pre-operatively
from a brain of one of the plurality of the living subjects.
[0019] The method further includes the steps of acquiring
electrophysiological information for one of the plurality of living
subjects, and relating the acquired electrophysiological
information to spatial coordinates in the brain image volume of the
corresponding living subject. Furthermore, the method includes the
step of registering the brain image volume of the corresponding
living subject to the common image volume of reference so as to
create an atlas in which any spatial coordinates of the brain of
the corresponding living subject are related to atlas coordinates
in the atlas such that the acquired electrophysiological
information associated with the related spatial coordinates in the
brain image volume of the corresponding living subject can be
related to atlas coordinates in the atlas, and vice versa.
Moreover, the method includes the step of storing the atlas in a
digitized format of files. The atlas has an architecture to be
accessible over a network.
[0020] Additionally, the method includes the steps of populating
the atlas with new electrophysiological information acquired from a
target of interest, accessing the electrophysiological information
from the atlas, and obtaining the electrophysiological information
from the atlas in one of a text format, an image format and a
mixture thereof, respectively. In one embodiment, the obtained
electrophysiological information includes an initial optimal target
position for at least one deep brain stimulator to be placed in a
brain of a target of interest.
[0021] In yet another aspect, the present invention relates to a
system that contains electrophysiological information related to at
least one of a plurality of living subjects, where any portion of
interest in the brain of one of the plurality of living subjects
and corresponding brain image volume may be identified by a set of
corresponding spatial coordinates.
[0022] In one embodiment, the system includes a data storage
device, and an atlas stored in the data storage device for
containing the electrophysiological information. The atlas is
created such that when a brain image volume is registered to the
atlas, any spatial coordinates of the brain image volume are
related to corresponding atlas coordinates in the atlas, and vice
versa. The atlas has an architecture to be accessible over a
network and is in communication with a user interface. The user
interface, in one embodiment, is used for populating the system
with new electrophysiological information acquired from a target of
interest, accessing the electrophysiological information from the
system, and obtaining the electrophysiological information from in
the system in one of a text format, an image format and a mixture
thereof, respectively. The system further includes a controller in
communication with the data storage device. The data storage device
has a memory.
[0023] In a further aspect, the present invention relates to a
computer readable medium or media. In one embodiment, the computer
readable medium or media includes a data structure and a user
interface in communication with the data structure. The data
structure relates to an atlas that contains electrophysiological
information related to at least one of a plurality of living
subjects, where any portion of interest for the brain of the at
least one of the plurality of living subjects and corresponding
image volume may be identified by a set of corresponding spatial
coordinates.
[0024] In one embodiment, the atlas is created such that when a
brain image volume of a living subject is registered to the atlas,
any spatial coordinates in the brain image volume of the living
subject are related to corresponding atlas coordinates in the
atlas, and vice versa. Therefore, the electrophysiological
information associated with spatial coordinates from which the
electrophysiological information is acquired in the brain of the
living subject can be related to atlas coordinates in the atlas,
and vice versa. The atlas is stored in a digitized format of
files.
[0025] The data structure, in one embodiment, includes a plurality
of transformations and corresponding inverses of the plurality of
transformations. Each of a plurality of transformations registers a
brain image volume to the atlas, where the brain image volume is
acquired from one of the plurality of living subjects.
[0026] In yet a further aspect, the present invention relates to a
method for optimal placement of a deep brain stimulator in a brain
of a target of interest. In one embodiment, the method includes the
step of acquiring from the target of interest pre-operatively at
least one piece of information associated with a state of brain
condition of the target of interest. The state of brain condition
is related to a type of a disease, and/or a degree of the disease.
The information associated with the state of brain condition of the
target of interest includes presenting complaints, locations of
symptoms related to one or more diseases, type and degree of the
one or more diseases, unified Parkinson's disease rating scale
scores both on and off medications, mini-mental status examination,
medications and dosages, cognitive performance, gait performance,
pre-operative target positions, and any mixture thereof of the
target of interest.
[0027] Furthermore, the method includes the steps of accessing
remotely an atlas that contains electrophysiological information
related to a plurality of living subjects, where any portion of the
brain of one of the plurality of living subjects and corresponding
brain image volume may be identified by a set of corresponding
spatial coordinates, entering the acquired information from the
target of interest to the atlas to find a match between the
acquired information and the electrophysiological information
contained in the atlas, and automatically obtaining an optimal
position in the brain of the target of interest for placing a deep
brain stimulator from the matched information. In one embodiment,
the atlas is stored in a memory device associated with a central
host computer. The accessing step includes the step of accessing
the atlas over a network from a client computer. The central host
computer and client computer are coupled to and in communication
with the network, respectively. The network, in one embodiment, has
at least one of a public network, a dedicated network, a local
network, and any combination of them. The public network, in one
embodiment, comprises the Internet.
[0028] Moreover, the method includes the steps of acquiring
electrophysiological information intra-operatively from the brain
of the target of interest and adjusting the optimal position in the
brain of the target of interest for placing a deep brain stimulator
accordingly. Additionally, the method includes the steps of finding
a final optimal position in the brain of the target of interest for
placing a deep brain stimulator from the adjusted optimal position
and placing a deep brain stimulator in the brain of the target of
interest. Also, in one embodiment, the method includes the steps of
downloading information related to the optimal position in the
brain of the target of interest for placing a deep brain stimulator
from the matched information to a local computer and adjusting the
optimal position in the brain of the target of interest for placing
a deep brain stimulator from the downloaded information. The method
further includes the step of updating the atlas from information
related to the target of interest.
[0029] In another aspect, the present invention relates to a method
for optimal placement of a deep brain stimulator in a brain of a
target of interest. In one embodiment, the method includes the step
of acquiring from the target of interest pre-operatively at least
one piece of information associated with a state of brain condition
of the target of interest. Furthermore, the method includes the
steps of accessing remotely an atlas that contains
electrophysiological information related to a plurality of living
subjects, wherein the atlas is formed with a plurality of clusters,
each cluster being related to a state of brain condition and having
a plurality of optimal positions for a deep brain stimulator
distributed therein, entering the acquired information from the
target of interest to the atlas to find a match between the
acquired information and the electrophysiological information
contained in the atlas, and automatically obtaining an optimal
position in the brain of the target of interest for placing a deep
brain stimulator from one of the plurality of optimal positions for
a deep brain stimulator distributed in the plurality of
clusters.
[0030] These and other aspects of the present invention will become
apparent from the following description of the preferred embodiment
taken in conjunction with the following drawings, although
variations and modifications therein may be affected without
departing from the spirit and scope of the novel concepts of the
disclosure.
BRIEF DESCRIPTION OF THE DRAWINGS
[0031] FIG. 1 schematically shows one example of a quadricpolar
deep brain stimulator that can be utilized to practice the present
invention.
[0032] FIG. 2 schematically shows a platform that can be utilized
to practice the present invention: (a) a perspective view of the
platform, and (b) a perspective view of the platform with a guiding
member in place.
[0033] FIG. 3 schematically shows a system that can be utilized to
practice the present invention.
[0034] FIG. 4 shows a post-operative CT image of a patient after
the bilateral DBS implantation according to one embodiment of the
present invention.
[0035] FIG. 5 shows images of the final DBS positions acquired
intra-operatively from a group of 8 patients with corresponding
atlas coordinates visually shown according to one embodiment of the
present invention: (a) a sagital view of the left side subthalamic
nucleus (hereinafter "STN") targets, (b) a transverse view of the
left side STN targets, (c) a coronal view of the left side STN
targets, (d) a sagital view of the right side STN targets, (e) a
transverse view of the right side STN targets, and (f) a coronal
view of the right side STN targets.
[0036] FIG. 6 shows images of the final DBS positions acquired
post-operatively from a group of 8 patients with corresponding
atlas coordinates visually shown according to one embodiment of the
present invention: (a) a sagital view of the left side STN targets,
(b) a transverse view of the left side STN targets, (c) a coronal
view of the left side STN targets, (d) a sagital view of the right
side STN targets, (e) a transverse view of the right side STN
targets, and (f) a coronal view of the right side STN targets.
[0037] FIG. 7 shows images of the final DBS positions acquired
intra-operatively from a group of 18 patients with corresponding
atlas coordinates visually shown according to one embodiment of the
present invention: (a) a sagital view of the left side and right
side STN targets, (b) a transverse view of the left side and right
side STN targets, and (c) a coronal view of the left side and right
side STN targets.
[0038] FIG. 8 is a flowchart showing a method for creating an atlas
containing electrophysiological information related to at least one
living subject according to one embodiment of the present
invention.
[0039] FIG. 9 is a flowchart showing a method for optimal placement
of a deep brain stimulator in a brain of a target of interest
according to one embodiment of the present invention.
[0040] FIG. 10 shows microelectrode recording (hereinafter "MER")
signals acquired from a target of interest and features extracted
from the MER signals according to one embodiment of the present
invention: (a) the MER signals at a position above the STN, (c) the
MER signals at the middle of the STN, and (e) the MER signals at a
position below the STN; and (b), (d), and (f), each showing an
interspike interval (hereinafter "ISI") histogram and feature
values corresponding to the MER signals of (a), (c), and (e),
respectively.
[0041] FIG. 11 shows mean burst index (hereinafter "BI")
color-coded and superimposed on MR images according to one
embodiment of the present invention: (a) a sagital view of the mean
BI, and (b) a coronal view of the mean BI.
DETAILED DESCRIPTION OF THE INVENTION
[0042] The present invention is more particularly described in the
following examples that are intended as illustrative only since
numerous modifications and variations therein will be apparent to
those skilled in the art. Various embodiments of the invention are
now described in detail. Referring to the drawings, like numbers
indicate like parts throughout the views. As used in the
description herein and throughout the claims that follow, the
meaning of "a," "an," and "the" includes plural reference unless
the context clearly dictates otherwise. Also, as used in the
description herein and throughout the claims that follow, the
meaning of "in" includes "in" and "on" unless the context clearly
dictates otherwise. Moreover, titles or subtitles may be used in
the specification for the convenience of a reader, which has no
influence on the scope of the invention. Additionally, some terms
used in this specification are more specifically defined below.
DEFINITIONS
[0043] The terms used in this specification generally have their
ordinary meanings in the art, within the context of the invention,
and in the specific context where each term is used.
[0044] Certain terms that are used to describe the invention are
discussed below, or elsewhere in the specification, to provide
additional guidance to the practitioner in describing various
embodiments of the invention and how to practice the invention. For
convenience, certain terms may be highlighted, for example using
italics and/or quotation marks. The use of highlighting has no
influence on the scope and meaning of a term; the scope and meaning
of a term is the same, in the same context, whether or not it is
highlighted. It will be appreciated that the same thing can be said
in more than one way. Consequently, alternative language and
synonyms may be used for any one or more of the terms discussed
herein, nor is any special significance to be placed upon whether
or not a term is elaborated or discussed herein. Synonyms for
certain terms are provided. A recital of one or more synonyms does
not exclude the use of other synonyms. The use of examples anywhere
in this specification, including examples of any terms discussed
herein, is illustrative only, and in no way limits the scope and
meaning of the invention or of any exemplified term. Likewise, the
invention is not limited to various embodiments given in this
specification.
[0045] As used herein, "around", "about" or "approximately" shall
generally mean within 20 percent, preferably within 10 percent, and
more preferably within 5 percent of a given value or range.
Numerical quantities given herein are approximate, meaning that the
term "around", "about" or "approximately" can be inferred if not
expressly stated.
[0046] As used herein, the term "living subject" refers to a human
being such as a patient, or an animal such as a lab testing
monkey.
[0047] As used herein, the term "target of interest" refers to a
living subject under treatment or test.
[0048] As used herein, "target," "position of interest," and
"target region" are synonyms in the specification and refer to an
object of stimulation in a deep brain of a living subject for
treatment of a brain-controlled disorder.
[0049] As used herein, "stimulation" refers to increase temporarily
the activity of a body organ or part thereof responsive to an input
signal to the body organ or part.
[0050] The terms "project," "map," and "transform," as used herein,
are synonyms in the specification and refer to a transformation of
a point of interest from a source image volume to a target image
volume, and vice versa.
[0051] The terms "place," "implant," and "insert," as used herein,
are synonyms in the specification and refer to put or embed a
device, such as a microelectrode recording lead, macrostimulation
lead, and/or a deep brain stimulator, into a target region of a
brain of a living subject.
OVERVIEW OF THE INVENTION
[0052] Optimal placement of a deep brain stimulator comprises an
iterative procedure and associated means for performing the task. A
target region of a brain of a target of interest is chosen
pre-operatively based on anatomical landmarks identified on MR
images. This target region or position is used as an initial
position that is refined intra-operatively using information at
least from one of microelectrode recordings and macrostimulation.
For example, first a microelectrode recording lead is placed into
the initial position to characterize neuronal firing patterns,
which are, in turn, used to infer locations of deep brain nuclei
relevant to the target region. A unipolar macrostimulation lead is
then placed into the inferred location to elicit responses in an
awake patient. Both of these sources of electrophysiological
information allow neurosurgeons, neurologists, and
neurophysiologists to establish functional boundaries and to
reconstruct a somatotopic organization of the structures of
interest so as to identify a final target position at which a deep
brain stimulator is to be placed. Because the length of the
procedure increases with the time it takes to adjust the DBS to its
final position, a good initial position is critical. On the other
hand, creation of an atlas that contains electrophysiological
information captured during the procedure for visualization of
these functional boundaries so as to identify target positions of
interest would gain great clinic relevance and facilitate the
process to find an optimal initial position for placement of a DBS
for a target of interest.
[0053] The present invention, in one aspect, relates to a method of
creating an atlas that contains electrophysiological information
related to at least one of a plurality of living subjects, where
any portion of the brain of one of the plurality of living subjects
and corresponding brain image volume may be identified by a set of
corresponding spatial coordinates.
[0054] Referring to FIG. 8, a representative flowchart 800 of the
method according to one embodiment of the present invention is
shown. At step 801, a brain image volume is chosen as a common
image volume of reference from a plurality of brain image volumes,
where each of the plurality of brain image volumes is acquired
pre-operatively from the brain of one of the plurality of the
living subjects. At step 803, electrophysiological information for
one of the plurality of living subjects is acquired. At step 805,
the acquired electrophysiological information is related to spatial
coordinates in the brain image volume of the corresponding living
subject. At step 807, the brain image volume of the corresponding
living subject is registered to the common image volume of
reference so as to create an atlas in which any spatial coordinates
of the brain of the corresponding living subject are related to
atlas coordinates in the atlas such that the acquired
electrophysiological information associated with the related
spatial coordinates in the brain image volume of the corresponding
living subject can be related to atlas coordinates in the atlas,
and vice versa. The atlas has an architecture to be accessible over
a network.
[0055] Furthermore, the method includes the step of storing the
atlas in a digitized format of files. Additionally, the method
includes the steps of populating the atlas with new
electrophysiological information acquired from a target of
interest, accessing the electrophysiological information from the
atlas, and obtaining the electrophysiological information from the
atlas in one of a text format, an image format and a mixture
thereof, respectively.
[0056] In another aspect, the present invention relates to a method
of optimal placement of a deep brain stimulator in a brain of a
target of interest by using an atlas that contains
electrophysiological information related to a plurality of living
subjects.
[0057] Referring to FIG. 9, a representative flowchart 900 of the
method according to one embodiment of the present invention is
shown. At step 902, at least one piece of information associated
with a state of brain condition of the target of interest is
acquired from the target of interest pre-operatively. The state of
brain condition is related to a type of a disease, and/or a degree
of the disease. At step 904, an atlas that contains
electrophysiological information related to a plurality of living
subjects is accessed remotely. The atlas is formed with a plurality
of clusters, where each cluster is related to a state of brain
condition and have a plurality of optimal positions for a deep
brain stimulator distributed therein. At step 906, the acquired
information from the target of interest is entered into the atlas
to find a match between the acquired information and the
electrophysiological information contained in the atlas. And at
step 908, an optimal position in the brain of the target of
interest for placing a deep brain stimulator is automatically
obtained from one of the plurality of optimal positions for a deep
brain stimulator distributed in the plurality of clusters.
[0058] These and other aspects of the present invention are more
specifically described below.
METHODS, IMPLEMENTATIONS AND EXAMPLES OF THE INVENTION
[0059] Patients and Data and Image Acquistions
[0060] In one embodiment of the present invention, a group of 8
patients who undergo deep brain stimulator implantation at a target
of the STN is chosen to gather a set of data for evaluating the
invented method. The group of patients and the number of patients
are employed merely as an example to acquire a set of data for
practicing the present invention, and the use of the group of
patients and the number of patients should not limit the scope of
the present invention. Each patient, was assigned a number from S1
to S8 as his or her identification. The data was collected after
obtaining an Independent Research Board (hereinafter "IRB")
approval at Vanderbilt University.
[0061] The set of data related to electrophysiological information
to be acquired for each patient is categorized in terms of
pre-operative information, intra-operative information and
post-operative information, respectively.
[0062] The pre-operative information in general includes at least
one piece of information associated with a state of brain condition
of a target of interest that is related to a type of a disease, and
a degree of the disease. Other information may also be acquired. In
one embodiment, the information associated with the state of brain
condition generally includes presenting complaints, locations of
symptoms related to one or more diseases, type and degree of the
one or more diseases, unified Parkinson's disease rating scale
scores both on and off medications, mini-mental status examination,
medications and dosages, cognitive performance, gait performance,
pre-operative target positions, and any mixture thereof. The
symptoms have at least one of upper extremity rigidity, lower
extremity rigidity, upper extremity dystonia and lower extremity
dystonia. The unified Parkinson's disease rating scale scores are a
rating tool for evaluating mentation, behavior and mood, activities
of daily living, motor activity, and complication of therapy for
the patient undergoing treatment.
[0063] The intra-operative information in general includes at least
one piece of information associated with at least one
microelectrode, where the information includes microelectrode
recordings, a position of the microelectrode recordings, a label of
a structure in which the microelectrode recordings is located, and
any mixture thereof. Other information may also be acquired. The
microelectrode recordings are characterized by a firing rate
(hereinafter "FR") that measures tonic activity and indices that
measures phasic activity, where the indices include a BI, a pause
ratio (hereinafter "PR"), a pause index (hereinafter "PI), and an
ISI histogram. The intra-operative information also includes at
least one piece of information associated with at least one
stimulation electrode, where the information includes voltages
applied to the at least one stimulation electrode, a response of
the patient undergoing treatment to the stimulation, differences in
voltage between disappearance of symptoms and appearance of side
effects, a position of the at least one stimulation electrode, a
final intra-operative target position of a deep brain stimulator to
be placed, and any mixture thereof. Other information may also be
acquired. The response of the patient undergoing treatment to the
stimulation includes loss of rigidity, location where the loss of
rigidity is observed, appearance of side effects, and/or location
affected by these side effects.
[0064] The post-operative information in general comprises at least
one piece of information associated with at least one deep brain
stimulator, where the information includes a position of the at
least one deep brain stimulator in post-operative computerized
tomographical images, optimal setting of the at least one deep
brain stimulator, overall assessment of the patient after placement
of the at least one deep brain stimulator, and any mixture thereof.
Other information may also be acquired.
[0065] At each stage of the deep brain stimulator implantation for
a patient undergoing treatment, the electrophysiological
information related to the patient is acquired either by devices
such as an image acquisition device and a data acquisition device,
and/or by a surgical team who is in charge of the treatment of the
patient. The image acquisition device is arranged, in use, to
acquire a CT image and/or a MR image for the patient. Any types of
clinic available CT imaging systems and MR imaging systems can be
used for patient image acquisition to practice the present
invention. In one embodiment, the data acquisition device has at
least one microelectrode and/or at least one stimulation electrode
placed in predetermined target positions of the brain of the
patient, for acquiring intra-operative information. The at least
one stimulation electrode, in one embodiment, includes a unipolar
macrostimulation lead. The data acquisition device further has at
least one deep brain stimulator placed in the final target position
of the brain of the patient for acquiring post-operative
information. Additionally, the data acquisition device has a
micropositioning drive for recording positions of the at least one
microelectrode, the at least one stimulation electrode and the
final target position for the at least one deep brain stimulator to
be placed, respectively. Implantation procedures of the at least
one deep brain stimulator for the patient and corresponding data
acquisitions according to one embodiment of the present invention
are detailed in the following sections.
[0066] All patients undergoing consideration for a DBS implantation
at the target of the STN are first evaluated by a neurologist
specializing in movement disorders, and their medications are
adjusted to optimize their condition. If patients reach advanced
Parkinsonian symptoms, such as rigidity, bradykinesia, tremor, and
dyskinesia, despite optimal medical therapy, they are considered
for the surgical therapy by a multi-disciplinary group involving
neurology, neurosurgery, neurophysiology, and neuropsychiatry
specialists. Target selection is decided upon by the surgical team
if no contraindications exist. A majority of patients with the
above symptoms are recommended for STN targeting of DBS
therapy.
[0067] Pre-operative target identification is performed by the
functional neurosurgeon and is based on an identification of the
AC-PC location seen on MR image (3D SPGR volumes, TR: 12.2 msec,
TE: 2.4 msec, voxel dimensions 0.85.times.0.85.times.1.3 mm.sup.3)
pre-operatively. For the STN target, a preliminary point is chosen
at 4 mm posterior, 12 mm lateral, and 4 mm inferior to the
mid-commissural point. The adjustments for the initial intended
target are made based on the width of the third ventricle and
anatomical asymmetries noted on the MR image scan, but these
adjustments usually have less than 1 mm deviations from the initial
intended target location.
[0068] In general, a target position of treatment for a target of
interest is selected by a surgical team, based on pre-operatively
acquired information that is associated with a state of brain
condition of the target of interest. The information associated
with the state of brain condition of the target of interest
comprises presenting complaints, locations of symptoms related to
one or more diseases, type and degree of the one or more diseases,
unified Parkinson's disease rating scale scores both on and off
medications, mini-mental status examination, medications and
dosages, cognitive performance, gait performance, pre-operative
target positions, and any mixture thereof of the target of
interest.
[0069] Traditional methodology for carrying out this stepwise
target localization and implantation procedure has been based on an
externally fixed, rigid fixture, called a stereotactic frame that
encompasses the patient's head and upon which the
micro-manipulating equipment can be mounted and maneuvered with
sub-millimetric precision. These various stereotactic frames have
been optimized to obtain accurate images used to create the initial
target trajectory and plan and then to reduce erroneous movement
associated with passage of the test electrodes and the final
implantation [5]. These frames typically require mounting the day
of surgery, subsequent imaging with either CT and/or MR image axial
slices, and target planning prior to starting the actual procedure
of intra-operative mapping and ultimate placement of the electrode
implantation into the final target.
[0070] Recently, a FDA approved miniature stereotactic frame,
called a Starfix platform (FHC Corporation, Bowdoinham, Me.), has
become clinically available. This device, also referred as a
platform hereafter, as shown in FIGS. 2A and 2B, allows for more
versatility with elective stereotactic procedures, such as DBS
implantation, and can be utilized to practice the present
invention. Other platforms including other stereotactic frames in
clinical use can also be used to practice the present invention.
The Starfix platform or frame is used in this particular embodiment
to relate spatial coordinates in the OR to spatial coordinates in
the images. The construction and/or use of the atlas do not depend
on the specific frame being used. The platform 200 has a platform
body 210, an adjustor 220 attached to the platform body 210 and a
plurality of legs 230 outwardly and equal-angularly extending from
the platform body 210. Each of the plurality of legs 230 has a hole
280 at an end portion for receiving a corresponding fiducial marker
post 240 implanted into the outer table of the skull of a patient
so as to secure the platform 210. The platform 210 also has a
guiding member 250. The guiding member 250 has a plurality of
guiding tubes 260 including a center guiding tube 270. The
positions of the guiding tubes 260 including the central tube 270
can be adjusted by the adjustor 220. The platform 210 is currently
manufactured as a customized tripod that can be mounted onto
bone-based fiducial marker posts 240. Each platform is uniquely
manufactured based on a stereotactically planned trajectory using
software designed to mathematically relate the location of such
bone markers with respect to brain structures [6]. The bone-based
fiducial marker having a fluid-filled cylinder that is visible on
both CT and MR images is detachably attached to a post that is
implanted into the outer table of the skull. These images can then
be used in the stereotactic software to designate a trajectory in
relation to the bone-based marker posts. The plan is sent to the
manufacturer who then translates the stereotactic plan into a
customized platform for a given trajectory through a rapid
prototyping facility. The resultant platform is shipped to the
hospital within a certain time frame and is used for mounting the
same types of micromanipulators that are used on traditional
stereotactic frames. The remaining portion of the procedure is the
same with respect to intra-operative localization of the final
target of implantation with the patient awake.
[0071] Each patient undergoing surgery receives either one (for
unilateral DBS implantation) or two (for bilateral DBS
implantation) platforms. Each leg of the platform is attached to a
corresponding bone-implanted post. For each patient, the
acquisition of data proceeds in three stages. First, under
anesthesia, the fiducial marker posts are implanted onto
predetermined positions on the skull of the patient, fiducial
markers such as Acustar.TM. fiducial markers (Z-Kat, Inc.,
Hollywood, Fla.) are attached to the posts. The use of this marker
and post in open craniotomies has been reported on earlier [6].
Other fiducial markers and posts can also be used to practice the
present invention. CT and MR image volumes are acquired by the
image acquisition device with the patient anesthetized and head
taped to the table to minimize motion. For examples, CT images
acquired at kvp=120 V, exposure=350 mas, 512.times.512 pixels
ranging in size from 0.49 to 0.62 mm, slice thickness=2 mm for one
patient, 1.3 mm for 2 patients, 1 mm for all others. MR images are
3D SPGR volumes, TR: 12.2, TE: 2.4, voxel dimensions
0.85.times.0.85.times.1.3 mm.sup.3 except for patient S7 for which
the voxel dimensions are 1.times.1.times.1.3 mm.sup.3. After image
acquisition, the fiducial markers are removed. With the help of
MR-CT registration software, for instance, VoXimo (FHC Corporation,
Bowdoinham, Me.), the surgeon selects the initial target points
based on AC-PC coordinates and associated entry points on the
surface of the skull. In addition, the centroids of the markers and
the directions of their posts are determined from the acquired
images. These data are sent electronically to a fabrication plant
where a customized platform is manufactured to fit the posts and
provide an opening positioned over the entry point and oriented
toward the target. These data are also saved in a data storage
device such as a memory to form a database. In one embodiment, the
data saved in the data storage device forms a relational database.
The data saved in the data storage device, in another embodiment,
forms an object oriented database, or a mixed object oriented
database. The database can be implemented in Microsoft Access
(Microsoft Corporation, Richmond, Wash.). The database can also be
implemented in other software, such as Oracle (Oracle Corporation,
Redwood Shores, Calif.), Microsoft SQL (Microsoft Corporation,
Richmond, Wash.) and IBM DB2 (IBM Corporation, Armonk, N.Y.).
[0072] Second, surgery begins with drilling a burr hole, for
instance, having 14 mm in diameter. Referring to FIG. 3, an adaptor
(not shown here) is attached to each post 352, the platform 350 is
attached to the adaptor, and a micropositioning drive 310 is
attached to the platform 350. In one embodiment, a
microTargeting.RTM. device (FHC Corporation, Bowdoinham, Me.) is
employed as the micropositioning drive, also referred as a
microdrive hereafter. A microelectrode recording lead is placed
into the patient at the selected initial target position through
the central tube of the guide member attached to the platform. The
position of the microelectrode recording lead is adjusted so as to
find a new position where resting firing frequencies are noted or
detected. The adjustment involves three-dimensional adjustment. In
addition to changes in depth, it is possible to re-insert a probe
320 along parallel tracks distributed within a 10 mm circle around
the initial track. The microelectrode lead is removed and an
unipolar macrostimulation lead is inserted to the new position as
determined by the microelectrode recordings. With the patient
awake, response to stimulation generated from the macrostimulation
lead is monitored as the position of the macrostimulation lead is
further adjusted until optimal stimulation to the deep brain target
is detected. When the final positions are selected, the
macrostimulation lead is removed and a deep brain stimulator lead
is inserted at the final position. Medtronic #3387 and #3389
quadripolar lead.RTM., as shown in FIG. 1 and described above can
be used as DBS lead to practice the present invention. Other types
of DBS lead can also be utilized to practice the present invention.
A DBS lead 100 is inserted to a depth such that the centroid 160 of
the four electrodes 110-140 of the DBS lead 100 is coincident with
the final position of the electrode on the unipolar
macrostimulation lead. The proximal end of the DBS lead is then
anchored to the skull and buried beneath the scalp. The platform is
then removed. Within twenty-four hours of surgery, the imaging
markers are re-attached to the posts and a post-operative CT scan
is acquired. If no complications occur, the patient is discharged
home within a day of the surgery. During the entire procedure
coordinates are read on the mircodrive. These physical coordinates
are transformed into pre-operative CT coordinates using the
software used for pre-operative planning. The CT image is then
registered to the corresponding MR image by the MR-CT registration
software, for instance, VoXim.RTM.. The MR image is then registered
to an atlas as described infra.
[0073] Third, within about two weeks or other time periods deemed
properly by the surgeon the patient is brought back to the
operating room and the DBS lead is attached to an internal pulse
generator, for example, a Soletra generator (Medtronic, Inc.,
Minneapolis, Minn.), under general anesthesia. This is usually done
as an outpatient procedure. Programming of the generators is
performed typically as an outpatient procedure one month later by a
neurologist.
[0074] To assess the final position of the DBS in the
post-operative CT scans, the centroid of the DBS contact-electrodes
needs to be detected in the CT images. Referring back to FIG. 1,
the DBS lead 100 includes four exposed platinum/iridium
contact-electrodes 110, 120, 130 and 140. The centroid 160 of the
DBS contact-electrodes is at midway between the inner two
contact-electrodes 120 and 130, which is the target point to which
the surgeon attempts to deliver stimulation. Referring to FIG. 4, a
post-operative CT image 400 of a patient after the bilateral DBS
implantation having two DBS leads 410 is shown. The wire leads 420
are running under skin from the DBS leads 410 to the internal pulse
generator (not shown).
[0075] For each patient, the MER signals are recorded
intra-operatively and saved using a recording device such as dual
channel LeadPoint system from Medtronic Neurological, Inc.,
Minneapolis, Minn. These signals are recorded along a
microelectrode path starting 10 mm above the pre-operative target
point and ending 5 mm below. Signals are recorded every 0.5 mm for
10 sec, and sampled at 22 KHz. After the procedure, the digitized
signals and corresponding positions are downloaded from the
LeadPoint system and stored on file in the data storage device for
use. For this particular example, 850 signal epochs have been
recorded.
[0076] For each patient, each piece of electrophysiological
information is related to spatial coordinates of the brain of the
patient from which the piece of electrophysiological information is
acquired. The acquired electrophysiological information is stored
in the data storage device for use as described above.
REGISTRATION ALGORITHMS AND ATLAS CREATION
[0077] An atlas is a common volume of reference in which any
spatial coordinates of a brain of a target of interest are related
to atlas coordinates in the atlas. Therefore, the acquired
electrophysiological information associated with the spatial
coordinates in the brain image volume of the target of interest can
be related to atlas coordinates in the atlas, and vice versa.
Creation of the atlas requires registering individual image volumes
to the common volume of reference, which corresponds to the spatial
normalization of each individual brain image. Two types of
registrations algorithms are needed to proceed with the creation of
the atlas: rigid and nonrigid registrations. The rigid registration
algorithm is employed to register a CT image volume to a MR image
volume of the same patient, while the non-rigid registration
algorithm is used to register the MR image to the common volume of
reference, and vice versa. This is needed because the
intra-operative positions of the recording and stimulating
electrodes are given in CT coordinates in a corresponding
pre-operative CT image. The registration between CT and MR image
volumes of the same patient can be implemented using a standard
mutual-information based algorithm as proposed by Maes et al. [12].
In one embodiment, the CT image is registered to the corresponding
MR image by MR-CT registration software, for instance, VoXim.RTM..
Two nonrigid registration algorithms developed at Vanderbilt
University are utilized hereto to register the MR image volume to
the common volume of reference. Other algorithms may also be
utilized to practice the present invention.
[0078] The first one is called a demon algorithm proposed by
Thirion [8]. The demons algorithm computes a transformation that
minimizes the voxel-by-voxel intensity difference between the
source image volume and the target image volume. This method is
itself derived from an instantaneous optical flow equation proposed
by Horn and Schunck [20] for motion tracking in image sequences (in
the present invention, the two image volumes to be registered are
viewed as two frames in a sequence). The basic assumption on which
this equation is based is that the image intensity value of a point
in the anatomy does not change as it is displaced. This permits the
computation of a velocity vector (or in the invention a
displacement vector) at each voxel that obeys the following
equation: 1 i x x t + i y y t + i z z t = - i t
[0079] in which i is the intensity value in the image at the point
with coordinates (x, y, z). This equation is under-constrained and
regularization techniques are used to smooth the displacement
field. Thirion proposes to decouple the computation of the
displacement field and its regularization as opposed to casting the
problem as one single optimization problem. The displacement at
each point in the image is first computed by solving the equation.
The displacement field is then regularized by filtering it with a
Gaussian filter. The larger the standard deviation of this filter
is, the smoother the displacement field is. The algorithm is
iteratively applied in a multi-scale way. The matching is first
computed on coarse downsampled images then successively to images
with a finer spatial resolution. This strategy has several
advantages: it speeds up the computations, improves the convergence
properties of the algorithm, and uses the fact that, for human
anatomy, macroscopic features are, in general, more stable than
microscopic features. In one embodiment of the present invention,
two image pyramids are derived from the images to be registered, up
to a predetermined scale. A number of iterations of the algorithm
are applied to the images at the coarsest scale and the results
obtained at this scale serve as initial conditions for the next one
until the finer scale is reached. Furthermore, an additional
mechanism such as a bijectivity constraint is used to ensure a
one-to-one correspondence between the two images to be matched.
Following the approach proposed by Burr [21] this is done by
computing both a direct and a reverse deformation field which are
maintained compatible such that T.sub.1.fwdarw.2{circle over
(.times.)}T.sub.2.fwdarw.1.congruent.I, with T.sub.1.fwdarw.2 the
deformation field from image 1 to image 2, T.sub.2.fwdarw.1 the
deformation field from image 2 to image 1, {circle over (.times.)}
indicating composition, and I the identity transformation.
[0080] This greatly increases the robustness of the algorithm, and
it has the advantage of insuring that both the forward, i.e, from
the reference volume to the individual volumes, and reverse, i.e.,
from the individual volumes to the reference volumes,
transformations are one-to-one.
[0081] In one embodiment of the present invention, another nonrigid
algorithm called an Adaptive Basis Algorithm (hereinafter "ABA")
[9] is developed, which operates on a quite different principle.
Rather than trying to minimize the intensity differences at every
voxel, this algorithm computes a transformation that maximizes the
Mutual Information (hereinafter "MI") between the images. In this
technique, inspired by the work of Rueckert et al. [10] and Meyer
et al [11], the deformation that registers one image (a source
image) onto the other (a target image) is modeled with a linear
combination of radial basis functions with finite support. The
similarity measure that drives the registration process is the
mutual information between the source image and the target image.
In this algorithm, several improvements over existing mutual
information-based non-rigid registration algorithm are implemented.
These include working on an irregular grid, adapting the compliance
of the transformation locally, decoupling a very large optimization
problem into several smaller ones, and deriving schemes to
guarantee the topological correctness of the transformations.
[0082] More specifically, the adaptive base algorithm includes the
following steps: at first, a source image and a target image are
defined to be one of the remaining N-1 image volumes and the atlas,
respectively. Second, an image pyramid for each of the source image
and the target image is created, respectively. Each image pyramid
has M levels. Each level of the image pyramid has a resolution and
is segmented with a corresponding scale so as to form a grid. Each
image pyramid is formed such that level i of the pyramid has lower
resolution and larger scale than level (i-1), where i=1, . . . , M,
and M is an integer greater than 1. Then, a deformation field,
v(x), x being a position vector, which registers the source image
volume to the target image volume, is defined, and the deformation
field is further initialized as v(x)=v.sub.M (x). In one
embodiment, the deformation field is initially set to be zero.
Furthermore, the deformation field, v.sub.i(x), is computed at
level i of the image pyramids, where the deformation field
v.sub.i(x) at level i is a sum of the deformation field at level
(i+1) and a linear combination of a set of radial basis functions
spaced on the grid of level i, so as to register the source image
volume to the target image volume at level i and where the
computing starts at level (M-1). Moreover, regions of
misregistration are identified, which is resulted from the step of
computing the deformation field v.sub.i(x) at level i. Additionally
each of the regions of misregistration are optimized independently
from each other by modifying the region of support and radial basis
functions corresponding to the region in the deformation field
v.sub.i(x). Furthermore, the computing step, the identifying step
and the optimal step are iterated at level (i-1) of the image
pyramids till level 1 is reached so as to incrementally construct a
final deformation field in the form of
v(x)=v.sub.1(x)+ . . . +v.sub.M(x).
[0083] The adaptive base algorithm further includes the step of
optimizing a constraint scheme for enforcing a Jacobian matrix of
the deformation field to remain uniformly invertible throughout a
domain of the source image volume and a corresponding domain of the
target image volume so as to generate topologically correct
transformations between the source image volume and the target
image volume.
[0084] To create an atlas, a MR image volume of a living subject is
chosen as a common image volume of reference. Registering MR image
volumes acquired from a target of interest to the common image
volume of reference will create an atlas. The image registration
can be utilized by a nonrigid registration algorithm, such as the
demon algorithm, the ABA algorithm, and others. Once the
transformation between one image volume and the atlas is computed,
any spatial coordinates in this volume, such as spatial coordinates
relating to the acquired electrophysiological information and the
spatial coordinates of DBS, can be transformed into corresponding
atlas coordinates in the atlas. In this example, a MR image volume
acquired from one of the group of 8 patients is chosen to serve as
the common volume of reference, where each of the group of 8
patients has at least one DBS implanted in the brain. Image volumes
acquired from the remaining 7 patients of the group of 8 patients
are respectively registered to the common image volume of reference
by a demon algorithm and/or the ABA algorithm so as to create the
atlas. Thus, there are at least 7 DBS atlas points (coordinators)
in the atlas, each is projected by the spatial point (coordinators)
of the at least one DBS in each of the image volumes for the
remaining 7 patients. The optimal DBS position in the atlas is
computed as the centroid of all the DBS positions after their
projection onto the atlas.
[0085] Predicting the initial optimal DBS position for each patient
is the inverse of the operation described above. It includes
projecting the optimal DBS position from the atlas to each
individual image volume. This does not require another registration
step because the transformation from the patient to the atlas and
from the atlas to the patient are computed simultaneously. The
nonrigid registration algorithms impose constraints on these
transformations to keep them almost inverse of each other to
produce bijective transformations. For instance, to predict an
initial optimal position in an image volume of a patent that the
deep brain stimulator is to be implanted using the atlas and
nonrigid registration algorithm, the image volume of the patent
needs being registered to the atlas by the nonrigid registration
algorithm so as to find a registration transformation of the image
volume to the atlas. Application of an inverse of the
transformation will project the optimal target position of the deep
brain stimulator in the atlas to the image volume so as to identify
the initial optimal position of the deep brain stimulator in the
targeted region of the brain of the patient.
VISUAL EVALUATION OF THE REGISTRATION RESULTS
[0086] Two examples of obtaining the DBS coordinates according to
the present invention are presented. The first one relies on
coordinates provided intra-operatively by a STarFix guidance system
during surgery. This system translates the physical coordinates of
the DBS electrode into pre-operative CT coordinates. The second one
relies on an algorithm to get the centroid of the deep brain
stimulator in the post-operative CT scans [22]. One can expect
differences between these coordinates, the causes of which are
several. First, the STarFix system is not perfectly accurate.
Second, the intra-operative target point is arrived at with a
microstimulating electrode. This electrode is then replaced by the
permanent DBS stimulator, which introduces the surgical placement
error. Third, the brain may shift during surgery because of
swelling and/or loss of cerebrospinal fluid. After surgery, the
brain returns to its normal state, which also causes the electrode
to move.
[0087] Referring now to FIGS. 5 and 6, first to FIG. 5, several
circles forms a cluster 510 and each circle in the cluster 510
corresponds to an atlas position of a final DBS target projected
onto the atlas 500. These final DBS targets, or more specifically,
spatial coordinators of each corresponding final DBS target, are
acquired intra-operatively by a STarFix guidance system. FIGS.
5(a), 5(b), and 5(c) show a sagital view, a transverse view, and a
coronal view of the left side DBS targets projected onto the atlas,
respectively, while FIGS. 5(d), 5(e), and 5(f) respectively show a
sagital view, a transverse view, and a coronal view of the right
side DBS targets projected onto the atlas. The results shown here
are obtained with the ABA algorithm. The results are qualitatively
similar with the demons algorithm. All spatial coordinates of the
individual deep brain stimulators in association with FIG. 5 have
been acquired intra-operatively by a STarFix guidance system.
[0088] Similar to FIG. 5 but with the spatial coordinates of the
DBS targets acquired with post-operative CT scans, FIG. 6
represents images of the final DBS spatial coordinates acquired
post-operatively from the group of 8 patients with corresponding
atlas coordinates visually shown according to one embodiment of the
present invention. In FIG. 6, several circles forms a cluster 610
and each circle in the cluster 610 corresponds to the atlas
position of a final DBS target projected onto the atlas 600. FIGS.
6(a), 6(b), and 6(c) show a sagital view, a transverse view, and a
coronal view of the left side DBS targets projected onto the atlas,
respectively, while FIGS. 6(d), 6(e), and 6(f) respectively show a
sagital view, a transverse view, and a coronal view of the right
side DBS targets projected onto the atlas.
[0089] Still referring to FIGS. 5 and 6, according to the present
invention, the positions of final DBS targets form a cluster in the
atlas (cluster 510 in FIG. 5 and cluster 610 in FIG. 6,
respectively) and it indicates that the atlas coordinates projected
from the spatial coordinates of the DBS positions acquired
intra-operatively form a tight cluster 510 than the cluster 610
formed by the atlas coordinates projected from the spatial
coordinates of the DBS positions acquired post-operatively.
[0090] According to another embodiment of the present invention,
images of atlas coordinates 700 that are transformed from the final
DBS positions acquired intra-operatively from a group of 18
patients are shown in FIG. 7. FIGS. 7(a), 7(b), and 7(c)
respectively show a sagital view, a transverse view, and a coronal
view of both the left side and the right side DBS targets. FIG. 7
clearly shows that the atlas coordinates projected from the spatial
coordinates of the DBS positions acquired intra-operatively from
the group of 18 patients also form a cluster 710 in each side of
the brain. Thus, according to the present invention, the position
of the optimal target may be a function of parameter related to a
state of brain condition of a patient such as disease type or
state. For example, patients who have prominent leg rigidity may
benefit from an implant centered in a cluster more posterior and
inferior in the STN target than someone with arm rigidity whose
ideal cluster may be more anterior and superior.
ATLAS-GUIDED INITIAL TARGET POSITIONS
[0091] Table 1 shows the atlas coordinates transformed from the
spatial coordinates of the final DBS positions for the eight
bilateral STN patients by using the ABA algorithms. The DBS
coordinates are acquired intra-operatively. Each patient, is
assigned a number from S1 to S8 in column Subject as his or her
identification. Columns Left and Right represent locations of the
bilateral DBS, that is, column Left corresponds to the left side
implantation of the DBS, while column Right corresponds to the
right side implantation of the DBS. Sub-columns X, Y and Z are
atlas coordinates of a DBS placed in a specific target region (left
side or right side) for a specific patient, which corresponds to an
individual point in the atlas, represented by a corresponding
circle 510 in FIG. 5. Rows S1 to S8 represent a set of atlas
coordinates of a bilateral DBS of patients S1 to S8, respectively.
For instance, Row S3 of Table 1 represents patient S3 having a
bilateral DBS implantation, where the atlas coordinates of the left
side DBS are (X, Y, Z)=(122.47, 107.80, 51.18) in unit of mm, and
the atlas coordinates of the right side DBS are (X, Y, Z)=(96.86,
107.02, 50.24) in unit of mm. The centroid of the atlas coordinates
of the DBS positions of the eight bilateral STN patients is
computed for the left side implantation and the right side
implantation, respectively, which are presented in Row Mean. Row
STD represents a standard deviation (hereinafter "STD") the atlas
coordinates relative to the centriod. Row SEM is a standard error
of the mean (hereinafter "SEM"). The Euclidean distance between
each point in the atlas and its corresponding centroid is
represented in the Dc column.
[0092] Similar to Table 1, Table 2 shows the atlas coordinates
transformed from the spatial coordinates of the final DBS positions
for the eight bilateral STN patients by using the demons
algorithms. The DBS coordinates are acquired intra-operatively.
1TABLE 1 Atlas coordinates transformed from spatial coordinates of
the final DBS positions that are acquired intra-operatively using
the ABA algorithms. Atlas Coordinates Transformed by the Adaptive
Basis Algorithm (in unit of mm) Sub- Left Right ject X Y Z Dc X Y Z
Dc S1 124.28 106.39 53.98 2.58 95.47 106.20 53.20 3.52 S2 121.88
106.30 53.44 1.44 N/A N/A N/A N/A S3 122.47 107.80 51.18 1.43 96.86
107.02 50.24 2.12 S4 119.79 106.27 51.78 2.74 101.38 104.91 50.23
3.40 S5 123.42 107.41 51.40 1.42 99.10 105.99 50.19 1.35 S6 123.42
107.99 53.21 1.82 96.68 104.69 52.96 2.57 S7 122.48 105.80 51.90
1.07 97.54 104.98 50.51 0.92 S8 121.70 106.60 50.91 1.52 99.85
104.69 49.63 2.34 Mean 122.43 106.82 52.22 1.75 98.12 105.49 51.00
2.32 STD 1.37 0.81 1.16 0.60 2.07 0.91 1.45 0.97 SEM 0.49 0.28 0.41
0.21 0.73 0.32 0.51 0.34
[0093]
2TABLE 2 Atlas coordinates transformed from spatial coordinates of
the final DBS positions that are acquired intra-operatively using
the demons algorithm. Atlas Coordinates Transformed by the Demons
Algorithm (in unit of mm) Left Right Subject X Y Z Dc X Y Z Dc S1
124.28 106.39 53.98 2.29 95.47 106.20 53.20 2.90 S2 121.60 104.24
54.43 1.94 N/A N/A N/A N/A S3 122.03 106.52 53.04 0.88 95.92 105.95
51.87 2.14 S4 120.31 106.73 51.97 2.54 99.78 105.72 51.23 2.21 S5
123.11 106.00 52.77 1.08 98.85 104.68 51.94 1.08 S6 122.42 105.63
54.60 1.32 97.01 103.79 53.97 2.44 S7 121.86 104.44 53.44 1.32
97.89 104.19 52.34 0.89 S8 122.07 105.63 52.16 1.15 99.92 104.58
49.61 3.22 Mean 122.21 105.70 53.30 1.56 97.83 105.01 52.02 2.12
STD 1.15 0.93 0.99 0.61 1.78 0.94 1.40 0.87 SEM 0.41 0.33 0.35 0.22
0.63 0.33 0.5 0.31
[0094] As shown in Tables 1 and 2, when the spatial coordinates of
the final DBS positions acquired intra-operatively are transformed
into the atlas coordinates by the ABA and Demons algorithms,
respectively, the STD and the SEM of the atlas coordinates have
small values, which indicate that the final positions of the DBSs
transformed into the atlas coordinates result in tight clusters. It
is also worth noting that even though these two algorithms are
based on very different similarity measures, they lead to
essentially identical results, suggesting that the
accuracy-limiting factor is not the registration algorithm used but
either the spatial resolution of the MR images, the accuracy of the
DBS positioning system, a bias introduced by the spatial
normalization scheme, normal inter-subject variation, suboptimal
intra-operative selection of the target, or a combination of
these.
[0095] Tables 3 and 4 show the same information as Tables 1 and 2
but with spatial coordinates of the deep brain stimulators that are
acquired from the post-operative CT scans. A comparison of Tables 1
and 2 with Tables 3 and 4 shows clearly that atlas coordinates of
the DBS positions do not cluster as well when the coordinates are
acquired post-operatively as when the coordinates are acquired
intra-operatively. For instance, the STD value for the left side
DBS in Table 1 is 0.60 mm, while the corresponding STD value in
Table 3 is 0.99 mm, which is 65% more than the STD value when the
coordinates are acquired intra-operatively. Similarly, the STD
value for the right side DBS in Table 1 is 0.97 mm, while the
corresponding STD value in Table 3 is 0.1.58 mm, which is 63% more
than the STD value when the coordinates are acquired
intra-operatively. The results shown in FIGS. 5 and 6 also indicate
the same trend, where the spatial coordinates of the DBS positions
are acquired intra-operatively in FIG. 5, and the spatial
coordinates of the DBS positions in FIG. 6 are acquired
post-operatively. The projected atlas coordinates of the DBS
positions form a tighter cluster 510 than the cluster 610 formed by
the projected atlas coordinates of the DBS positions
accordingly.
[0096] The distance between an individual atlas point projected
from a corresponding DBS position and the centroid of atlas
coordinates projected from DBS positions of eight patients is
significantly smaller for the intra-operative coordinates than that
for the post-operative CT coordinates. Statistical significances
for one-sided t-tests are as follows: ABA algorithm for a left side
STN target with P<0.03, Demons algorithm for a left side STN
target with P<0.01, ABA algorithm for a right side STN target
with P<0.01, Demons algorithm for a right side STN target with
P<0.01, where P, with a value ranging from zero to one,
corresponds to a probability value of a statistical hypothesis test
as known to people skilled in the art.
3TABLE 3 Atlas coordinates transformed from spatial coordinates of
the final DBS positions that are acquired with post-operative CT
scans using the ABA algorithm. Atlas Coordinates Transformed by the
Adaptive Basis Algorithm (in unit of mm) Sub- Left Right ject X Y Z
Dc X Y Z Dc S1 123.01 107.65 53.81 1.69 95.38 107.6 54.73 4.11 S2
121.44 107.14 55.25 2.87 N/A N/A N/A N/A S3 123.88 108.15 49.84 3.3
97.59 110.55 48.76 4.95 S4 117.64 107.39 52.37 4.36 102.89 106.25
52.65 4.67 S5 123.18 107.35 53.14 1.38 99.18 106.72 51.3 1.09 S6
121.27 108.85 53.14 1.64 93.86 105.6 55.34 5.72 S7 123.48 105.57
51.9 2.53 97.13 105.16 50.5 2.5 S8 122 108.18 50.29 2.27 102.1
105.89 50.35 4.22 Mean 121.99 107.54 52.47 2.5 98.3 106.82 51.95
3.9 STD 2 0.97 1.79 0.99 3.33 1.83 2.41 1.58 SEM 0.71 0.34 0.63
0.35 1.18 0.65 0.85 0.56
[0097]
4TABLE 4 Atlas coordinates transformed from spatial coordinates of
the final DBS positions that are acquired with post-operative CT
scans using demons algorithm. Atlas Coordinates Transformed by the
Demons Algorithm (in unit of mm) Sub- Left Right ject X Y Z Dc X Y
Z Dc S1 123.01 107.65 53.81 1.74 95.38 107.6 54.73 3.4 S2 121.44
104.86 55.42 2.65 N/A N/A N/A N/A S3 123.34 107.13 51.63 2.37 96.98
108.94 50.04 4.02 S4 118.54 108.08 52.62 3.76 100.79 107.41 53.52
3.08 S5 122.76 106 54.46 1.51 98.97 105.68 52.97 1.19 S6 120.59
106.68 53.91 1.4 94.08 104.35 56.14 5.41 S7 122.9 104.15 53.56 2.54
97.5 104.37 52.36 2.09 S8 122.2 107.04 51.3 2.15 102.08 105.93
50.75 4.67 Mean 121.85 106.45 53.34 2.26 97.97 106.33 52.93 3.41
STD 1.62 1.36 1.4 0.76 2.86 1.73 2.13 1.46 SEM 0.57 0.48 0.5 0.27
1.01 0.61 0.75 0.52
[0098] The tighter the cluster of the projected atlas points of the
DBS positions in the atlas is, the better the results are.
Therefore, using the intra-operative coordinates according to the
present invention to predict an initial target position may lead
better results than using the post-operative CT coordinates.
Nevertheless, as described above, both the intra-operative DBS
coordinates and the post-operative DBS coordinates can be utilized
to practice the present invention. One explanation for this
discovery is that the spread of the clusters in the atlas increases
if measurement noise in the DBS positions used to create this atlas
also increases. The sources of error associated with the
post-operative DBS coordinates involve the centroid detection
algorithm and the errors associated with registering pre-operative
and post-operative CT images. The difference is the surgical
placement error, i.e, the distance between the target point chosen
intra-operatively and the position of the permanent DBS.
5TABLE 5 Distances between the initial target positions selected
manually and the final DBS positions acquired intro-operatively and
distance between the initial target positions selected
automatically and the final DBS positions acquired
intro-operatively (the atlas used herein has been generated with
the intro-operative DBS coordinates). Target Prediction Errors
intra-operative Atlas (in unit of mm) Left Right Automatic
Automatic Subject Manual ABA Demons Manual ABA Demons S1 5.95 2.58
2.4 6.94 3.52 2.94 S2 5.72 1.78 3.39 N/A N/A N/A S3 2.53 2.41 1.5
4.49 2.45 2.23 S4 5.3 2.46 2.8 1.99 3.24 2.36 S5 2.31 2.17 1 3.64
1.6 1.51 S6 5.95 2.36 2.85 7.31 3.39 3.84 S7 2 1.64 2.37 2.01 1.75
1.68 S8 1.71 1.75 0.7 1.67 2.94 3.67 Mean 3.93 2.14 2.13 4.01 2.7
2.6 STD 1.94 0.37 0.96 2.36 0.78 0.91 SEM 0.69 0.13 0.34 0.83 0.28
0.32
[0099] Thus, according to the present invention, when projected
onto a common reference volume, optimal DBS positions result in
tight clusters if these positions can be determined accurately in
each individual patient. These results also show, albeit in an
indirect way, that the coordinates acquired intra-operatively are
more accurate than the coordinates acquired post-operatively, and
suggest a high accuracy for use of a platform from which
intro-operative coordinates of the final DBS are acquired.
[0100] Tables 5 exhibits the Euclidean distances between the
manually selected initial target positions and the final DBS
positions and the Euclidean distances between the automatically
selected initial target positions and the final DBS positions,
respectively, for each patient. The final DBS positions are
acquired intro-operatively. The atlas used in this case has been
generated with the intro-operative DBS coordinates. The Euclidean
distance between an initial target position and a final DBS
position represent a target prediction error.
[0101] Specifically, column Subject of Table 5 in represents the
group of 8 patients with each patient having a number from S1 to S8
assigned as his or her identification. Columns Left and Right of
Table 5 represent locations of the bilateral DBS, that is, column
Left corresponds to the left side implantation of the DBS, while
column Right corresponds to the right side implantation of the DBS.
Sub-column Manual of Table 5 represents the Euclidean distance
between the manually selected initial target position and the final
DBS position in a specific side implantation of the DBS. For
example, sub-column Manual of column Left corresponds to the
Euclidean distance in the left side implantation of the DBS, while
sub-column Manual of column Right corresponds to the Euclidean
distance in the right side implantation of the DBS. And sub-column
Automatic of Table 5 having sub sub-columns ABA and Demons
represents the Euclidean distance between the automatically
selected initial target position and the final DBS position in a
specific side implantation of the DBS. The automatically selected
initial target position is computed based on the ABA algorithm for
sub sub-column ABA, while it is computed based on the Demons
algorithm for sub sub-column Demons. Rows S1 to S8 of Table 5
represent a set of the Euclidean distances (or target prediction
errors) of patients S1 to S8, respectively. For instance, Row S3 of
Table 5 represents patient S3 having a bilateral DBS implantation,
where the target prediction errors for the left side DBS
implantation are 2.53 mm for manually selected initial target
position, 2.41 mm for automatically selected initial target
position with the ABA algorithm, and 1.5 mm for automatically
selected initial target position with the Demons algorithm,
respectively, and the target prediction errors for the right side
DBS implantation are 4.49 mm for manually selected initial target
position, 2.45 mm for automatically selected initial target
position with the ABA algorithm, and 2.23 mm for automatically
selected initial target position with the Demons algorithm,
respectively. Row Mean represents a mean value of the target
prediction errors. Row STD represents a standard deviation of the
target prediction errors relative to the mean value of the target
prediction errors. And row SEM is a standard error of the mean.
[0102] Similar to Table 5 but with the atlas generated with the
post-operative DBS coordinates, Table 6 represents the Euclidean
distances between the initial target positions selected manually
and the final DBS positions acquired intro-operatively and the
Euclidean distances between the initial target positions selected
automatically and the final DBS positions acquired
intro-operatively, respectively, for each patient.
6TABLE 6 Distances between the initial target positions selected
manually and the final DBS positions acquired intro-operatively and
distances between the initial target positions selected
automatically and the final DBS positions acquired
intro-operatively (the atlas used herein has been generated with
the post-operative DBS coordinates). Target Prediction Errors
Post-operative Atlas (in unit of mm) Left Right Automatic Automatic
Subject Manual ABA Demons Manual ABA Demons S1 5.95 1.69 1.74 6.94
4.11 3.4 S2 5.72 2.87 3.69 N/A N/A N/A S3 2.53 5.05 3.64 4.49 5.67
5.23 S4 5.3 3.71 4.16 1.99 3.55 2.54 S5 2.31 2.02 2.05 3.64 1.31
1.42 S6 5.95 2.41 2.2 7.31 6.56 6.27 S7 2 4.12 4.07 2.01 4.22 3.7
S8 1.71 2.89 2.74 1.67 4.46 5.45 Mean 3.93 3.09 3.04 4.01 4.27 4
STD 1.94 1.13 0.97 2.36 1.66 1.73 SEM 0.69 0.4 0.34 0.83 0.59
0.61
[0103] Table 5 demonstrates that for the data sets obtained and
used as described herein, an atlas-guided placement of DBS is not
only feasible but also is better than the technique in current
clinical use. With both ABA and demons registration algorithms, the
initial target points are substantially closer to the final ones
than the initial target point chosen manually. It is shown that the
average distance between an initial position selected with the
automatic method of the present invention and a final position of a
DBS is 45% smaller on the left side and 30% on the right,
respectively, than the one between an initial position selected
manually and a final position of a DBS. For the group of 8
patients, for example, in Table 5, the mean distances between the
initial targets and final positions are 2.14 mm (ABA algorithm) and
2.70 mm (ABA algorithm) on the left and right sides, respectively,
for the atlas-based automatic method, as compared with 3.93 mm and
4.01 mm for the manual method.
[0104] Despite the small size of the data sets employed in the
study, the distance between the initial target points and the final
target points is significantly smaller (P<0.01, one sided paired
t-test) than the distance between the initial target points chosen
manually and the final target points for both ABA and demons
algorithms on the left side. On the right side, the significance is
only slightly smaller (P<0.07) and (P<0.06) for the ABA and
demons algorithms, respectively. A comparison of Tables 5 and 6
also reveals that when using the post-operative CT coordinates to
create the atlas, atlas-guided placement of DBS does not do much
better than the current manual approach. This is consistent with
what is described above in connection with Tables 1 to 4 that
tighter clusters are formed with the intra-operative coordinates
than that formed with the post-operative CT coordinates.
7TABLE 7 Distances between the initial target positions selected
manually and the final DBS positions acquired intro-operatively and
distance between the initial target positions selected
automatically and the final DBS positions acquired
intro-operatively (the atlas used herein has been generated with
the intro-operative DBS coordinates acquired from a group of 18
patients). Target Prediction Errors ultra-operative Atlas (in unit
of mm) Left Right Subject Manual Automatic Manual Automatic P1 2.53
2.36 4.49 3.17 P2 1.61 2.75 5.03 4.31 P3 5.45 0.83 N/A N/A P4 5.12
2.54 N/A N/A P5 0 0.97 5.8 2.9 P6 1.71 1.67 1.67 1.63 P7 3.14 3.19
0.48 1.91 P8 1 1.13 2.08 2.56 P9 3.48 2.92 3.27 2.93 P10 1.99 3.34
1.99 3.52 P11 2.31 2.24 3.64 2.16 P12 0 0.71 2 2.34 P13 1.5 2.93
N/A N/A P14 1.51 4.72 N/A N/A P15 5.95 3.75 7.31 2.18 P16 5.95 2.37
6.94 2.98 P17 2.01 1.38 3 2.54 P18 5.85 5.22 5.38 4.18 Mean 2.7 2.5
3.88 2.79 STD 1.98 1.32 2.04 0.77
[0105] Similar to Table 5 but for a group of 18 patients, Table 7
presents the Euclidean distances between the manually selected
initial target positions and the final DBS positions acquired
intro-operatively and the Euclidean distances between the
automatically selected initial target positions and the final DBS
positions acquired intro-operatively, respectively. The atlas used
in this case is generated with the intro-operative DBS coordinates.
In Tables 7, each patient is assigned a number from P1 to P18 in
column Subject as his or her identification.
[0106] For the group of 18 patients, Table 7 again displays that
the atlas-projected initial target point is substantially closer to
the final ones than the initial target point chosen manually for a
target of interest, that is, the atlas-based automatic target
localization approach is better than the manual one. For example,
the mean distances between the initial targets and final positions
are 2.50 mm and 2.79 mm on the left and right side, respectively,
for the atlas-based automatic method, as compared with 2.7 mm and
3.88 mm for the manual method.
[0107] By comparing Table 5 with Table 7, one notes that the mean
distances between the atlas-predicted initial targets and final DBS
positions, and the standard deviation of the error are different.
For example, the mean distances and the standard deviation for the
group of 18 patients are about 2.5 mm and 1.32 mm for the left side
DBS implementation, and 2.79 mm and 0.77 mm for the right side DBS
implementation, respectively. However, for the group of 8 patients,
the mean distances and the standard deviation are about 2.14 mm and
0.37 mm for the left side DBS implementation, and 2.7 mm and 0.78
mm for the right side DBS implementation, respectively.
DATA POPULATING AND REPRESENTING
[0108] In the exemplary embodiment as described supra, the method
of creating an electrophysiological atlas is illustrated with a
group of 8 patients undergoing treatment. The method is further
extended to a large group having 18 patients. According to the
present invention, the atlas is created with an architecture such
that new electrophysiological information acquired can be populated
into the atlas and electrophysiological information contained in
the atlas can be accessed via a user interface. For instance, after
a set of image volumes (pre-operative MR and CT volumes) of a
target of interest has been acquired, the CT volume is registered
to the MR image, which is then registered to the atlas. The
corresponding computed transformations are saved on files in a
memory device and used to project patient coordinates onto atlas
coordinates, and vice versa. The memory device can be associated
with a central host computer which itself can be in communication
with a local and/or a public network such as the Internet.
[0109] Based on pre-operatively acquired information associated
with a state of brain condition of the target of interest, the
surgical team makes a pre-operative surgical plan. The state of
brain condition is related to a type of a disease, and/or a degree
of the disease. The information associated with the state of brain
condition of the target of interest generally includes presenting
complaints, locations of symptoms related to one or more diseases,
type and degree of the one or more diseases, unified Parkinson's
disease rating scale scores both on and off medications,
mini-mental status examination, medications and dosages, cognitive
performance, gait performance, pre-operative target positions, and
any mixture thereof of the target of interest.
[0110] From a client (or local) computer, in communication with the
central host computer, the surgical team accesses remotely or
locally the atlas stored in the central host computer, and enters
the acquired information from the target of interest to the atlas
to find a match between the acquired information and the
electrophysiological information contained in the atlas. Then an
optimal position in the brain of the target of interest for placing
a deep brain stimulator from the matched information is
automatically selected from the atlas. In one embodiment, the
automatically selected optimal position is given in terms of an
electrode path for a single DBS implantation (or two electrode
paths for a bilateral DBS implantation). A cylindrical region
around the electrode path is defined in the atlas accordingly.
[0111] The entire database (atlas) is queried, and smaller
databases that contain selected information about the points within
these cylinders are created. In one embodiment, the entire database
including the smaller databases is accessed by the surgical team
over a network from a client computer during the intra-operative
DBS implantation. The network can be one of a public network such
as the Internet, a dedicated network, a local network, and any
combination of them. The public network includes the Internet. In
another embodiment, these smaller databases are downloaded to a
client (or local) computer together with the corresponding
transformations for the target of interest for the surgical use of
the intra-operative DBS implantation. This approach is favored over
accessing the entire database across the network to avoid potential
problems with network connections during the surgery and to permit
real-time searches on the local computer that can be brought into
the operating room. In an alternative embodiment, the entire
database including the smaller databases is written into a computer
readable medium or media, such as, compact discs (CD), floppy
disks, hard drivers, and the likes, for the use of the
intra-operative DBS implantation in an operating room where the
network may not be available.
[0112] In one embodiment, the intra-operative positions of the
electrodes, for example, a microelectrode recording lead and a
macrostimulation lead, placed in the brain of the target of
interest are acquired by a microdrive, such as a
microTargeting.RTM. device. The microdrive is in communications
with the local computer in the operating room, and/or the central
host computer through the network for uploading the intra-operative
positions of the electrodes. The acquired positions are
automatically transformed into atlas coordinates via a pre-computed
registration. This approach permits intra-operative queries such as
"show me all the points in the atlas with a mean firing frequency
within a given range and within 5 mm of the current position of the
electrode". This query returns a list of records, each of which is
a pointer to a file that contains a series of positions that are
then color-coded and displayed on the patient image volume. This
approach can be used, for instance, to assess whether the firing
rate which the surgical team is measuring is a likely value at the
current electrode position, and whether the measured firing rate is
observed at a distance away from the current position, which would,
in turn, indicate whether the electrode is at the anatomic
position.
[0113] Referring now to FIG. 10, MER signals are acquired along an
electrode path at different positions from a patient, and features
associated with the MER signals, including a FR, a BI, a PI, a PR,
and an ISI histogram, are extracted. As shown in FIGS. 10a, 10c and
10e, signals 1010, 1012 and 1014 are raw MER signals acquired at a
position above the STN, at the middle of the STN and at a position
below the STN, respectively. Spike trains 1011, 1013 and 1015 are
respectively extracted from raw MER signals 1010, 1012 and 1014.
ISI histograms 1020, 1022 and 1024 associated with the
corresponding raw MER signals 1010, 1012 and 1014 are respectively
shown in FIGS. 10b, 10d and 10f. For each of raw MER signals 1010,
1012 and 1014, the FR, the BI, the PI, and the PR are also
extracted. For example, for raw MER signals 1010, FR=44.0,
BI=1.047, PI=0.505, PR=6.215, as shown in FIG. 10b, for raw MER
signals 1012, FR=58.8, BI=0.871, PI=0.378, PR=2.213, as shown in
FIG. 10d, and for raw MER signals 1014, FR=117.7, BI=1.462,
PI=0.092, PR=0.385, as shown in FIG. 10f.
[0114] Once these features have been extracted, their values can be
color-coded and displayed in the atlas. FIG. 11 shows the mean
value of the BI in the electrophysiological atlas for a patient
undergoing treatment, which is color-coded. Bright and dark pixels
correspond to high and low values of the BI, respectively. One can
distinguish several regions with low, medium, and high values for
this feature. Low values correspond to white matter, medium values
to the STN, and high values to structures such as the niagra
(hereinafter "Ni") and the Vim (another nucleus). For instance, as
shown in FIG. 11, region 1110 is the STN, region 1120 is the Ni,
and region 1130 represents the Vim, respectively. Although the
scarcity of data does not yet permit a precise localization of
complete nuclei boundaries, the results obtained according to the
present invention clearly show patterns in the data and clusters
that correspond to known anatomical structures that are invisible
in MR images.
FURTHER OBSERVATIONS AND DISCUSSIONS
[0115] In the present invention, among other things, a method for
creating an atlas containing electrophysiological information
related to at least one living subject is disclosed, which can be
used for automatic selection of the pre-operative target for DBS
placement and identification of boundaries of structures and
substructures which are not visible in current imaging modalities
for intra-operative guidance. This is achieved by correlating
intra-operative recordings with electrophysiological information
contained in the atlas, which permits the surgical team to identify
the current location of the electrode, and plan and execute
displacements from this position. The current difficulty is the
intra-operative acquisition of signals that cover a region around
the various targets of interest. Recording equipment used in
current clinical practice only permit recording of one channel at a
time. Very recently, a 10 channel recording device developed by FHC
Corporation, Bowdoinham, Me., has been clinically available. This
device permits recording along 5 parallel tracks on each side for a
total of 10 simultaneous channels. This may allow the signal
database of the present invention to be rapidly expanded which, in
turn, may improve the localization of substructure boundaries based
on electrophysiological signatures.
[0116] Additional improvements may be made. Because the number of
patients for gathering the necessary data is limited, the method
has been evaluated on the set of data used to create the atlas.
This may bias the results in the favor of the method. As the number
of data sets increase, a separation of the image volumes into
training and testing set may address this issue. Other approach may
be to use a synthesized average image as the atlas. Additionally,
all the image volumes are employed in the study regardless of
clinical outcome. An alternative approach may be to select only
cases for which the clinical outcome is excellent to build the
atlas. However, none of these issues affect the utilization of the
present invention.
[0117] While there has been shown several and alternate embodiments
of the present invention, it is to be understood that certain
changes can be made as would be known to one skilled in the art
without departing from the underlying scope of the invention as is
discussed and set forth above and below. Furthermore, the
embodiments described above are only intended to illustrate the
principles of the present invention and are not intended to limit
the scope of the invention to the disclosed elements.
[0118] List of References
[0119] [1]. Referen G. Deuschl, J. Volkmann, and P. Krack, "Deep
brain stimulation for movement disorders", Movement Disorders, vol.
17 (supplement 3), pp S 1 -S 1, 2002.
[0120] [2]. B. Schrader, W. Hamel, D. Weinert, and H. M. Mehdorn,
"Documentation of electrode localization." Movement Disorders, vol.
17 (supplement 3), pp S167-S174, 2002.
[0121] [3]. J. L. Vitek, Mechanisms of deep brain stimulation:
excitation or inhibition. Movement Disorders, vol. 17 (supplement
3), pp S69-S72, 2002.
[0122] [4]. A. M. Lozano, Deep brain stimulation for Parkinson's
disease. Vol. 7, no. 3, pp 199-203, 2001.
[0123] [5]. R. L. Galloway and R. J. Maciunas, "Stereotactic
neurosurgery", Crit Rev Biomed Eng, vol. 18, no.3, pp 181-205,
1990.
[0124] [6]. J. Franck, P. Konrad, R. Franklin, F. Haer, and D.
Hawksley. "STarFix: A Novel Approach to Frameless Stereotactic
Neurosurgery Utilizing a Miniaturized Customized Pretargeted
Cranial Platform Fixture--Technical Description, Unique Features,
and Case Reports", Movement Disorders Society, 7th Intl. Congress
of Parkinsons Disease & Movement Disorder, Miami, Fla.,
November 2002.
[0125] [7]. C. R. Maurer, Jr., J. M. Fitzpatrick, M. Y. Wang, R. L.
Galloway, Jr., R. J. Maciunas, and G. S. Allen, "Registration of
head volume images using implantable fiducial markers," IEEE Trans.
Med. Imaging, vol. 16, pp 447-462, 1997.
[0126] [8]. J. P. Thirion, "Image matching as a diffusion process:
an analogy with Maxwell's demons". Medical Image Analysis, vol. 2,
no. 3, pp 243-260, 1998.
[0127] [9]. G. Rhode, A. Aldroubi and B. M. Dawant, "The
Adaptive-bases algorithm for intensity-based nonrigid image
registration," IEEE Transactions on Medical Imaging, vol. 22, no.
11, pp 1470-1479, 2003.
[0128] [10]. D. Rueckert, L. I. Sonoda, C. Hayes, D. L. G. Hill, M.
0. Leach, and D. J. Hawkes, "Nonrigid Registration Using Free-Form
Deformations: Application to Breast MR Images." IEEE Transactions
on Medical Imaging, vol. 18, no. 8, pp 712-721, 1999.
[0129] [11]. C. R. Meyer, J. L. Boes, B. Kim, P. Bland, K. R.
Zasadny, P. V. Kison, K. Koral, K. A. Frey, and R. L. Wahl.,
"Demonstration of accuracy and clinical versatility of mutual
information for automatic multimodality image fusion using affine
and thin-plate" Medical Image Analysis, vol. 3, pp 195-206,
1997.
[0130] [12]. F. Maes, A. Collignon, and P. Suetens, "Multimodality
image registration by maximization of mutual information," IEEE
Transaction on Medical Imaging vol. 16, no. 2, pp 187-198,
1997.
[0131] [13]. J. D. Atkinson, D. L. Collins, G. Bertrand, T. M.
Peters, G. B. Pike, and A. F. Sadikot, "Optimal location of
thalamotomy lesions for tremor associated with Parkinson Disease: a
probabilistic analysis based on postoperative magnetic resonance
imaging and an integrated digital atlas", J. Neurosurgery, vol. 96,
pp 854-866, 2002.
[0132] [14]. G. Schaltenbrand and W. Wahren, Atlas for Stereotaxy
of the Human Brain. Stuttgart, Germany: Thieme, 1977.
[0133] [15]. K. W. Finnis, Y. P. Starreveld, A. G. Parrent, A. F.
Sadikot, and T. M. Peters, "Threedimensional database of
dubcortical dlectrophysiology for dmage-guided stereotactic
functional neurosurgery", IEEE Transactions on Medical Imaging,
vol. 22 (11), pp 93-104, 2003.
[0134] [16]. J. Talairach and P. Toumeau, Co-Planar Stereotaxic
Atlas of the Human Brain. Stuttgart, Germany: Georg Thieme Verlag,
1988.
[0135] [17]. P. St-Jean, A. F. Sadikot, D. L. Collins, D. Clonda,
R. Kasrai, A. C. Evans, and T. M. Peters, "Automated atlas
integration and interactive 3-dimensional visualization tools for
planning and guidance in functional neurosurgery," IEEE Trans. Med.
Imag., vol. 17, pp 672-680, 1998.
[0136] [18]. G. Deuschl, J. Volkmann, P. Krack, "Deep brain
stimulation for movement disorders", Movement Disorders, vol. 17
(supplement 3) pp, S 1-S 1, 2002.
[0137] [19]. Deuschl, G., et al., "Deep brain stimulation of the
subthalamic nucleus for Parkinson's disease: a therapy approaching
evidence-based standards." J Neurol, 2003. 250 Suppl 1: p.
I43-146.
[0138] [20]. B. Horn and B. Schunck, "Determining optical flow",
Artificial Intelligence, vol. 17, pp.185-203, 1981.
[0139] [21]. D. J. Burr, "A dynamic model for image registration."
Computer Graphics and Image Processing, vol. 15, pp. 102-112,
1981.
[0140] [22]. C. Nickele, E. Cetinkaya, J. Michael Fitzpatrick, and
P. E. Konrad. "Method for Placing Deep-Brain Stimulators",
Proceedings of Medical Imaging 2003: Image Processing, SPIE, (in
press).
* * * * *