U.S. patent application number 11/116667 was filed with the patent office on 2006-11-02 for method of determining lung condition indicators.
Invention is credited to Richard G. Caro.
Application Number | 20060243280 11/116667 |
Document ID | / |
Family ID | 37214383 |
Filed Date | 2006-11-02 |
United States Patent
Application |
20060243280 |
Kind Code |
A1 |
Caro; Richard G. |
November 2, 2006 |
Method of determining lung condition indicators
Abstract
A method and a microprocessor controlled device for use in
medical evaluation of the pulmonary system. The method and device
incorporate two different non invasive evaluative techniques.
Passive auscultation provides indicators of lung condition. Active
auscultation, a different evaluative technique, provides
information or indications generally not available or provided from
passive auscultation. Combining the information from both
techniques greatly improves the ability to provide effective
medical treatment.
Inventors: |
Caro; Richard G.; (San
Francisco, CA) |
Correspondence
Address: |
PARSONS HSUE & DE RUNTZ LLP
595 MARKET STREET
SUITE 1900
SAN FRANCISCO
CA
94105
US
|
Family ID: |
37214383 |
Appl. No.: |
11/116667 |
Filed: |
April 27, 2005 |
Current U.S.
Class: |
128/204.23 |
Current CPC
Class: |
A61B 5/085 20130101;
A61B 8/08 20130101; A61B 7/003 20130101 |
Class at
Publication: |
128/204.23 |
International
Class: |
A61M 16/00 20060101
A61M016/00 |
Claims
1. A method of determining the condition of a patient's lungs,
comprising: making an active measurement of the patient's lungs by
injecting a signal comprising audible frequencies through the
patient's lungs; making a passive measurement of the patient's
lungs by monitoring the breath sounds of the patient during a
breathing cycle; combining results from the active measurement with
the passive measurement.
2. The method of claim 1 wherein making an active measurement
further comprises determining one or more variables from the
measurement.
3. The method of claim 2 wherein one variable is the velocity of
the injected signal.
4. The method of claim 3 wherein the velocity is determined over a
range of frequencies.
5. The method of claim 4 wherein the range of frequencies comprises
a range from about 20 HZ to about 50 kHZ.
6. The method of claim 2 wherein one variable is the phase of the
injected signal.
7. The method of claim 2 wherein one variable is the dispersion of
frequencies over a range of frequencies.
8. The method of claim 1 further comprising ceasing to make an
active measurement while making a passive measurement.
9. The method of claim 1 further comprising calculating a ratio of
an amplitude of the signal at a first frequency versus the
amplitude of the signal at a second frequency.
10. The method of claim 1 further comprising calculating a ratio of
a velocity of the signal at a first frequency versus a velocity of
the signal at a second frequency.
11. The method of claim 1 further comprising calculating a ratio of
an attenuation of the signal at a first frequency versus the
attenuation of the signal at a second frequency.
12. A method of determining the condition of a patient's lungs,
comprising: injecting a signal including audible frequencies
through the patient's lungs; recording the injected signal and
processing the signal to determine a first set of criteria
indicative of a condition of the lungs; monitoring naturally
occurring breath sounds of the patient's lungs; processing the
breath sounds of the patient's lungs to determine a second set of
criteria indicative of a condition of the lungs; and narrowing the
number of criteria indicative of a condition of the lungs by
comparing the first set of criteria from the injected signal with
the second set of criteria from the naturally occurring breath
sounds.
13. The method of claim 12 wherein recording the injected signal
comprises receiving the injected signal with a plurality of
transducers.
14. The method of claim 12 wherein monitoring the naturally
occurring breath sounds comprises monitoring a plurality of
transducers.
15. The method of claim 12 wherein recording the injected signal
does not occur at the same time as monitoring the naturally
occurring breath sounds.
16. A device for evaluating the condition of a lung, comprising: an
emitting transducer that injects a signal having audible
frequencies through the lung; one or more receiving transducers
that receive the injected signal; and a processing unit that
monitors the injected signal, and determines a first set of
indicators of the condition of the lung from the injected signal,
and that monitors naturally occurring breath sounds of the lung and
determines a second set of indicators of the condition of the lung,
and that compares the first and second set of indicators and
provides a third set of indicators comprising the overlap of a
portion of the first and second indicators respectively, thereby
providing a set of indicators of lung condition based upon both the
injected signal and the naturally occurring breath sounds.
17. A device for evaluating the condition of a lung, comprising: an
emitting transducer that injects a signal having audible
frequencies through the lung; one or more receiving transducers
that receive the injected signal, the one or more receiving
transducers also monitoring natural breath sounds of the lung; and
means for producing an indication of the condition of the lung
based upon both the natural breath sounds and upon the injected
signal.
18. A method of determining the condition of a patient's lungs,
comprising: injecting a signal comprising audible frequencies into
the patient's lungs; determining the speed of sound of the injected
audible signal through the patient's lung tissue; calculating a
property of the lung tissue; correlating the property with a first
indication of the condition of the patient's lungs; monitoring an
amplitude of a breath sound of the patient's lungs as the patient
inhales and exhales; correlating the monitored breath sound
amplitude with a second indication of the condition of the
patient's lung; and combining the first and second indications to
narrow the possibilities of the condition of the patients
lungs.
19. The method of claim 18 where the condition comprises one or
more of emphysema, bronchitis, chronic obstructive pulmonary
disorder, or asthma.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] The present application is related to U.S. application Ser.
No. ______, of Wilkinson et al., entitled "Method And Apparatus For
Determining Conditions Of Biological Tissues," and filed on ______
and U.S. application Ser. No. 10/272,494 of Wilkinson et al., filed
on Oct. 15, 2002, which is a continuation of Patent Cooperation
Treaty Application No. PCT/AU01/00465, filed on Apr. 20, 2001,
which claims priority to Australian Provisional Application Nos. AU
PQ7040 and AU PR4333, filed on Apr. 20, 2000 and Apr. 10, 2001,
respectively. U.S. application Ser. No. 10/272,494, was published
as U.S. Patent Application Publication No. 2003/0120182 A1. The
present application is also related to U.S. application Ser. No.
______, of Wilkinson et al, filed on Apr. 21, 2005 as a
continuation of aforementioned U.S. application Ser. No.
10/272,494, and entitled "Apparatus And Method For Lung Analysis."
The present application is also related to Australian Application
Nos. AU 2001252025 and 2004222800, filed on Apr. 20, 2001 and Oct.
4, 2004, respectively. Each of the foregoing applications,
provisional applications, and publications, is hereby incorporated
herein, in its entirety, by this reference.
FIELD OF THE INVENTION
[0002] The present invention relates to a method and apparatus for
determining characteristics of biological tissues in humans and
animals. In particular, it relates to determining the
characteristics of tissues such as the lungs and airways.
BACKGROUND OF THE INVENTION
[0003] Non-invasive determination of the condition of the pulmonary
system is useful in determining proper medical treatment.
[0004] Ultrasonic waves have been used to monitor and observe the
condition of patients and of selected tissues. However, ultrasonic
techniques are not very effective in tissues in which there is a
substantial quantity of gas, such as the lung.
[0005] Respiratory problems ail infants and adults alike. Among
infants, respiratory failure is a common problem requiring support
and is usually treated with a period of mechanical ventilation.
Over the last decade the mortality of infants suffering respiratory
failure has shown an impressive decline and the vast majority of
infants now survive initial acute respiratory illness, but lung
injury associated with mechanical ventilation causes many infants
to develop `chronic lung disease.` Chronic lung disease is
characterised by persisting inflammatory and fibrotic changes.
Adults are often afflicted with different respiratory diseases or
conditions. Some common lung diseases or conditions include
emphysema, asthma, regional collapse (atelectasis), interstitial
oedema and both focal lung disease (e.g. tumour) and global lung
disease (e.g. emphysema).
[0006] A clear need exists for a simple, non-invasive and
convenient system to monitor and assess the condition of the
lung.
SUMMARY OF THE INVENTION
[0007] The present invention provides a non invasive method and
apparatus for use in determination of lung condition.
[0008] One aspect of the invention is a method of determining the
condition of a patient's lungs. The method comprises making an
active measurement of the patient's lungs by injecting a signal
comprising audible frequencies through the patient's lungs, making
a passive measurement of the patient's lungs by monitoring the
breath sounds of the patient during a breathing cycle, and
combining results from the active measurement with the passive
measurement.
[0009] Another aspect is a method of determining the condition of a
patient's lungs. The method comprises injecting a signal including
audible frequencies through the patient's lungs and then recording
the injected signal and processing the signal to determine a first
set of criteria indicative of a condition of the lungs. The method
also comprises monitoring naturally occurring breath sounds of the
patient's lungs and processing the breath sounds of the patient's
lungs to determine a second set of criteria indicative of a
condition of the lungs. With these two sets of criteria, the number
of criteria indicative of a condition of the lungs is narrowed by
comparing the first set of criteria from the injected signal with
the second set of criteria from the naturally occurring breath
sounds.
[0010] Yet another aspect of the invention is a device for
evaluating the condition of a lung. The device comprises an
emitting transducer that injects a signal having audible
frequencies through the lung, one or more receiving transducers
that receive the injected signal, and a processing unit that
monitors the injected signal. The processing unit determines a
first set of indicators of the condition of the lung from the
injected signal. The processing unit also monitors naturally
occurring breath sounds of the lung and determines a second set of
indicators of the condition of the lung. It then compares the first
and second set of indicators and provides a third set of indicators
comprising the overlap of a portion of the first and second
indicators respectively, thereby providing a set of indicators of
lung condition based upon both the injected signal and the
naturally occurring breath sounds.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] A reference numeral, letter, and/or symbol that is used in
one drawing to refer to a particular element or feature may be used
in another drawing to refer to a like element or feature.
[0012] FIG. 1 is schematic illustration of system 100.
[0013] FIG. 2(a) illustrates an electric circuit which models the
acoustic characteristics of the thorax.
[0014] FIG. 2(b) illustrates (1) large, (2) moderate and (3) small
acoustic losses as measured using the electric circuit model and
which represents the output sound pressure level as would be
measured at the chest.
[0015] FIGS. 3A-3D are graphs of some measured and derived
indicators associated with active auscultation.
[0016] FIG. 4 is a chart of normal and adventitious respiratory
sounds.
[0017] FIGS. 5A-5C are graphs of signals associated with passive
auscultation.
[0018] FIG. 6 is a graph of velocity as a function of
frequency.
[0019] FIG. 7 is a graph of attenuation as a function of frequency
and disease states that may be indicated.
[0020] FIG. 8 is a graph of wheeze amplitude as a function of
crackle amplitude.
[0021] FIG. 9 is a three dimensional representation of the
multidimensional analysis of the present invention.
DETAILED DESCRIPTION
[0022] The present invention utilizes a combination of two
different techniques to aid in determination of the condition of
the lungs. The first technique is passive auscultation, which
involves listening to the natural breath sounds of the lungs as a
patient inhales and exhales (during inspiration and expiration).
Passive auscultation goes back to the early 1800's when the
stethoscope and associated techniques were developed. Of course,
modem electronics-provide for much more sophisticated measurement
and analysis. The second technique is active auscultation which
involves actively introducing a signal in the lungs and monitoring
the signal after it passes through some portion of the lungs.
Unlike ultrasonic techniques, this utilizes a lower frequency
spectrum comprising audible frequencies.
[0023] An understanding of the theoretical aspects of sound
transmission in tissue is helpful for the best use of bio-acoustic
data which is obtained using the present invention.
[0024] The term "auscultation" is commonly used and well known in
medical circles. Herein, the concept or technique commonly known
simply as "auscultation" is referred to as "passive auscultation"
in order to distinguish it from an "active auscultation" concept or
technique associated with the present invention, as further
described below.
Active Auscultation
[0025] The term "active auscultation" generally refers to actively
introducing at least one audible signal into the body or a portion
thereof and thereafter receiving and/or monitoring the signal after
it has passed through some portion of the body. This is done to aid
in the diagnosis and/or treatment of the body or a portion thereof.
The audible signal that is introduced to the body or portion
thereof may be selected to suit its particular application, and
signal parameters can be tailored to the application. Examples of
the parameters include but are not limited to: the time of the
introduction of the sound or any of various parameters of the
sound, such as the sound pressure level, the phase of the sound,
the frequency of the sound, the velocity of the sound, and the
like, for example, such that the relative nature or condition of
the second derivative or responsive sound may be analyzed in a
meaningful way, such as quantitatively, for example.
[0026] In general, active auscultation may involve the
cross-correlation of the injected signal as it is introduced to the
body or a portion thereof and the signal as it is received after
passing through some portion of the body, be it transmitted,
reflected, scattered, refracted, and/or the like, and obtaining
meaningful information from the correlation, such as a time delay
or a phase shift, etc. The information obtained may concern a
single parameter, such as a sound velocity, for example, multiple
parameters, such as a sound velocity and a sound attenuation, for
example, and/or a ratio of parameters, such as a ratio of a first
sound velocity and a second sound velocity, for example, as further
described herein.
[0027] According to the present invention, any of various
parameters of the received signal may be determined. A
consideration of a single parameter may be useful in assessing or
determining a condition of a body or a portion of a body. Examples
of such single parameters include an amplitude, a pressure, a
velocity, a frequency, an attenuation, a phase, a time, and the
like, associated with the injected and/or received signal. As will
be discussed later, consideration of multiple parameters is of even
greater utility and preferable.
[0028] A unique feature of audible signal propagation through the
lung parenchyma is that the velocity is less than that expected for
either tissue (1500 ms.sup.-1) or air (343 ms.sup.-1). This can be
explained, in part, by examining the basic relationship between
audible signal velocity .nu. and the physical properties of the
lung tissue through which the audible signal is propagating. This
relationship is: v = 1 .rho. .times. .times. C , ( 1 ) ##EQU1##
where .rho. is the density and C is the volumetric compliance or
inverse volumetric stiffness per unit volume. In determining the
velocity of an audible signal in air, substituting an air density
of 1.2 kgm.sup.-3 and an air compliance of 7.14.times.10.sup.-6
Pa.sup.-1 yields a audible signal velocity in air of 342
ms.sup.-1.
[0029] It has been shown that this relationship also holds for
composite porous materials with a closed cell structure which is
similar to that of the lung, but where .rho. and C are replaced by
the tissue's average or composite values. For more information
please refer to Rice, D. A. (1983) Sound speed in pulmonary
parenchyma. J. Appl. Physiol. 54:304-308, which is hereby
incorporated by this reference in its entirety. Expressing these
values in terms of the volumetric fraction of tissue h and of gas
(1-h) and the constituent densities and compliances gives tissue
density: .rho.=(1-h).rho..sub.g+h.rho..sub.t (2), and volumetric
compliance: C=(1-h)C.sub.g+hC.sub.t (3), where
.rho.,.rho..sub.g,.rho..sub.t are the composite, gas and tissue
densities respectively and C,C.sub.g,C.sub.t are the composite, gas
and tissue volumetric compliances respectively.
[0030] Substituting equations (2) and (3) into equation (1) yields
an expression which relates audible signal velocity through a
composite structure to the volumetric fraction and the physical
properties of both the tissue and gas which compose the material: v
= 1 ( ( 1 - h ) .times. .rho. g + h .times. .times. .rho. t )
.times. ( ( 1 - h ) .times. C g + hC t ) . ( 4 ) ##EQU2##
[0031] It must also be noted that the density of air is
approximately 3 orders of magnitude less than that of most tissues
and the volumetric compliance of air is some 4 orders of magnitude
larger than that of most tissues. This can be used to determine the
velocity of audible signal propagation through the lung for a range
of volumetric fractions which are likely to be seen in the lung,
(0.05 at TLC to 0.5 to 0.9 for a fully atelectatic/collapsed lung).
These velocities can be determined by simplifying equation 4 as
follows: v = 1 h .function. ( 1 - h ) .times. 1 .rho. t .times. C g
. ( 5 ) ##EQU3##
[0032] Equation 5, in combination with FIG. 3 illustrates the
dependence that audible signal velocity has on the volumetric
fraction of tissue, the volumetric fraction of air, the tissue
density and the gas compliance. The tissue compliance and the gas
density play essentially no role in the determination of
velocity.
[0033] Audible signal velocity in composite materials is determined
in part by the product of the tissue density and the gas
compliance. The result of this is that the lung parenchyma appears
to act like homogeneous mass-loaded air as far as audible signal
propagation is concerned, such that the velocity of audible signal
propagation through the tissue is markedly slower than through air.
Substitution of known values for tissue density, .rho..sub.t and
gas compliance, C.sub.g in equation 5 gives: v = 11.82 h .function.
( 1 - h ) . ( 6 ) ##EQU4##
[0034] Differentiation of .nu. in equation 6 with respect to h
determines a minimum value for velocity at h=0.5 where .nu.=23.6
ms.sup.-1. For values of h<0.5 the velocity increases with
decreasing lung density and conversely for h>0.5 the velocity
decreases with decreasing lung density. This is clarified by way of
illustration in FIG. 3.
[0035] The quadratic properties of equation 6 result in the
presence of two values for h for any particular value of measured
velocity. These values are: h=0.5.+-. {square root over
(0.25-139.56/.nu..sup.2)} (7).
[0036] Therefore, the determination as to whether h is above or
below 0.5 must be made on physical grounds or by making paired
velocity measurements where h is changed between measurements. The
direction of the associated change in velocity (increasing or
decreasing) can then be used to indicate whether h is above or
below 0.5. Therefore, the volumetric fraction of tissue and gas in
the lung and hence lung density can be determined directly from
measuring the velocity of an audible signal as it propagates
through the tissue. All of this theoretical explanation is given
only to inform the reader on the underpinnings of auscultation and
signal propagation. It should be understood that the present
invention should not be limited in any way to use of these specific
formulae and that calculations and processing techniques can be
implemented in any number of ways.
[0037] The audible signal may be introduced in any non-invasive
manner, such as by percussion, or using any mechanical, electrical
or other transducer which is capable of generating acoustic sounds.
It is preferable that the audible signal which is introduced to the
tissue possesses properties which allow it to easily be
distinguished from environmental noise which may be present.
Examples may include a single tone or a sinusoidal wave. In a
preferred embodiment of the invention, a pseudo-random noise is
produced by an electro-acoustic transducer and introduced into the
tissue. Looking at FIG. 1, the signal is produced by sending
transducer 122 of measurement system 100. Sending transducer is
placed in contact with the patient, preferably at some distance
from signal receiver 116, such that the signal must pass through
the lungs or other biological tissue that is being measured.
Although one sending transducer is shown in FIG. 1, a plurality of
sending transducers may alternatively be utilized. Signal receiver
116 has one or more receiving transducers 118. Preferably more than
one transducer 118 is employed, and the transducers 118 can be
arranged in any geometry. In the preferred embodiment shown, the
transducers 118 are arranged in various locations within a support
medium 120, which is preferably pliant or anatomically shaped and
allows the transducers to maintain contact with the patient, either
directly or though some thin sterile material.
[0038] The sending transducer(s) 122 and signal receiver 116 are
respectively connected to control unit 110 through wires 114 and
112 respectively. Wireless transmission may alternatively be used
in place of wires 114 and 112. Control unit 110 comprises processor
110A, display 110B, storage 110C, and human interface devices 110D.
These are all connected by system bus 111 and other circuitry.
Human interface devices may include a keyboard and mouse etc . . .
and display 110B may also be a touch screen type display. Various
other circuitry (not shown) is included in control unit 110 for
signal generation and processing. Control unit 110 may be a purpose
built device or may be implemented within a personal computer with
additional circuitry contained in a separate unit or insertable
card.
[0039] It is preferable to use a pseudo-random noise signal that
has characteristics which are similar to a white noise signal, but
with mathematical properties which allow its amplitude to be
defined at any moment in time. Furthermore, it is preferred that
introduction of the pseudo-random noise signal to the tissue occurs
in bursts, preferably of 0.1 to 20 seconds duration, and the
signals are produced preferably with frequencies which range from
20 Hz to 25 kHz and at a sound pressure level of between 1 and 100
Pascal.
[0040] The output of signal receiver 116 may be amplified using low
noise isolation amplifiers and band-pass filtered with cut-off
frequencies and roll-off characteristics which depend on the
acoustic properties of the tissue which is being assessed. For
example, for measurements made on the neonatal lung, the pass band
is preferably between 50 Hz and 5 KHz with a roll-off which
corresponds to that of a 4.sup.th order linear phase filter. These
filters remove any very low frequency environmental noise (e.g.
below 10 Hz) that can adversely affect the performance of
auto-scaling amplifiers into which the filtered signal may be
fed.
[0041] The output signal from signal receiver 116 may be amplified
by control unit 110 and is then processed. Processing includes a
cross-correlation analysis of the input and output signals.
[0042] The cross-correlation function can be calculated using the
output of any of sending transducers 122 as the input signal, x(t)
and the output of any of the transducers 118 located on the other
side of the tissue as the output signal, y(t) wherein the
cross-correlation function can be calculated as: R xy .function. (
.tau. ) = lim T .fwdarw. .infin. .times. 1 T .times. .intg. 0 T
.times. x .function. ( t ) .times. y .function. ( t + .tau. )
.times. .times. d t , ##EQU5## where T is the observation time, and
.tau. is the delay time between x(t) and y(t) at which
R.sub.xy(.tau.) is calculated.
[0043] The impulse response of the system in the time domain can
also be determined. It is preferable that the impulse response then
undergoes Fast Fourier Transformation so that the signal is
transformed into the frequency domain and the transfer function of
the tissue can be determined. This transfer function provides a
quantitative indication of the characteristics of the tissue,
wherein:
[0044] (a) the magnitude of the transform provides data relating to
the transmission of the audible signal as it propagates through the
tissue as a function of frequency; and
[0045] (b) the phase of the transform (after "unwrapping") can be
used to calculate the phase difference, time delay and velocity of
the audible signal for each frequency that is present in the
psuedo-random noise signal which is introduced to the tissue by the
acoustic transducer. Each of these may serve as an indicator of the
condition of the lung.
[0046] A separate analysis of the relative transmission of the
audible signal through the tissue can be used to identify resonant
and anti-resonant frequencies of the tissue which is being
assessed. Changes in these frequencies can then be used to assess
regional differences in tissue topology which may be related to
pathology. These are also examples of indicators of lung
condition.
[0047] Despite numerous experimental investigations of
trans-pulmonary audible signal transmission where the source of
audible signal is placed at the mouth, there has been no
theoretical model which described audible signal transmission
through the thorax. The present invention may, but does not
necessarily, employ a simple model based on the double wall
transmission model that is used in architectural acoustics to
describe the audible signal attenuating effect of double walls
separated by a compliant air layer, as is present in the lung.
[0048] The main features of this model as it relates to the thorax
can be represented by an electrical equivalent circuit that can be
used to describe the pertinent features of audible signal
transmission through the thorax. This model is illustrated in FIG.
2(a). This approach to the analysis of acoustic transmission across
the thorax facilitates analysis using sophisticated circuit
emulation software such as SPICE to explore the effect of changing
model parameters. In the equivalent electric circuit model
where:
[0049] R.sub.cw is the loss component associated with the chest
wall and parenchyma;
[0050] M.sub.cw, M.sub.P is the surface mass of the chest wall and
parenchyma respectively;
[0051] C.sub.gl is the lung gas compliance;
[0052] P.sub.in, P.sub.o are the acoustic input and output audible
signal pressure levels respectively; and [0053] R.sub.0 is the
acoustic impedance of free space (414 MKS Rayls).
[0054] As illustrated in FIG. 2(b), the model can be used to
simulate the effect that changing R.sub.cw has on the transfer
function of the equivalent circuit which represents the chest. This
transfer function can be described mathematically as
P.sub.o(f)/P.sub.in(f) where f is the frequency and P.sub.in(f) and
P.sub.o(f) are the input (transducer) and receiving transducer
sound pressure levels (SPL) respectively. As R.sub.cw is decreased,
the transfer function becomes progressively more peaked or resonant
as illustrated by curves 1 to 3 in FIG. 4(b).
[0055] At sufficiently high frequencies, the output sound pressure
level for all three curves falls asymptotically at a rate of 60 dB
per decade. As the frequency is increased above the resonant
frequency, the response is dominated by the inertial mass of the
proximal and distal chest walls, and the shunt gas compliance of
the lung. These act together to produce the 60 dB per decade
fall-off, such that the thorax is, in effect, acting like a third
order low-pass electrical filter. Analysis of the equivalent
circuit, neglecting losses, shows that the resonant frequency of
the thorax, f.sub.0, can be determined using: f 0 = 1 2 .times.
.pi. .times. 2 C gl .function. ( M cw + M p ) . ( 8 ) ##EQU6##
[0056] Furthermore, if the transfer function is measured at f.sub.0
and at another frequency well above f.sub.0, say, 3f.sub.0 then
using an analysis of the equivalent circuit, an explicit expression
for lung gas compliance, C.sub.gl, can be deduced in the form C gl
= 4.18 .times. 10 - 2 .times. G f 0 , ( 9 ) ##EQU7## where
G=|P.sub.o(f)/P.sub.in(f)| and is the magnitude of the transfer
function of the thorax measured at 3f.sub.o.
[0057] It follows that gas volume V.sub.gl can be computed using
equation 10: V.sub.gl=.gamma.P.sub.0C.sub.gl (10) where .gamma. is
the adiabatic gas constant and P.sub.0 is the atmospheric
pressure.
[0058] An important component of acoustic transmission which can be
modelled using the equivalent electric circuit is the loss
component Rcw illustrated in FIG. 2(a) which includes acoustic loss
in the chest wall and parenchyma. Because the chest wall is
acoustically thin, the dissipative loss in the wall is negligible
but the loss in the parenchyma, which includes a large number of
serial mass-compliance interfaces formed from the tissue and gas
comprising the parenchymal structure, may be considerable. One
model that has been proposed to account for acoustic loss in the
parenchyma comprises air bubbles in water, for which an analysis
already exists. In this model, absorption occurs because acoustic
work is required to alternately compress and expand these
bubbles.
[0059] It has been shown that the plane wave attenuation produced
by N bubbles over distance x is given by: P .function. ( x ) = P 0
.times. e - ( N .times. .times. .sigma. 2 ) .times. x , ( 11 )
##EQU8##
[0060] where
.sigma.=16.pi..sup.2r.sub.o.sup.4.rho..sub.tc.sub.tR/{R.sup.2+(.omega.M-1-
/.omega.C).sup.2}; [0061] P(x) is the SPL at x; [0062] P.sub.o is
the SPL at x=0; [0063] r.sub.0 bubble radius; [0064] c.sub.t sound
speed in tissue; and [0065] R,M,C are the effective mechanical
resistance, mass and compliance of the bubbles respectively.
[0066] Attenuation, .alpha. = P .function. ( x ) P 0 ##EQU9## in
dB/cm can then be written as: .alpha.=4.35N.sigma. (12).
[0067] This is a complex function of R,M,C but a simplified
expression for the attenuation can be deduced by recognizing that
the acoustic vibration of the bubbles (alveoli) is dominated by
bubble compliance C at frequencies which are much lower than
resonance (i.e. <.apprxeq.10 kHz for realistic alveoli sizes).
Therefore, attenuation can be reduced to:
.alpha.=2.36.times.10.sup.-2r.sub.0.sup.6f.sup.3N (13).
[0068] The number of bubbles per unit volume N is approximately
related to the gas fraction (1-h) by: N = 3 .times. ( 1 - h ) 4
.times. .pi. .times. .times. r 0 3 , ( 14 ) ##EQU10## hence
equation 13 can be written as .alpha. = 1.35 .times. 10 - 3 .times.
f 3 .function. ( 1 - h ) 2 N . ( 15 ) ##EQU11##
[0069] From these equations, it can be seen that:
[0070] (a) absorption is related to the square of the gas fraction
(1-h); a small increase in the tissue fraction h is associated with
a marked decrease in high frequency attenuation (FIG. 6). This may
explain the increased transmission of audible signal across the
chest wall which can be observed clinically at high frequencies,
following pneumonic consolidation of the lung; and
[0071] (b) attenuation is a strong function of both the frequency f
and the alveolar radius r.sub.0. This may explain, in part, the
rapid fall-off in transmitted audible signal at high frequencies
seen in both adult and neonatal subjects. The dependence on bubble
radius may explain the reduced transmission through the thorax
during emphysema.
[0072] Furthermore, these equations indicate that:
[0073] (a) absorption is related to the square of the gas fraction
(1-h); and
[0074] (b) audible signal transmission attenuation is s strong
function of both the frequency and the alveolar radius.
[0075] Using these relationships between audible signal
transmission velocity in tissues and the tissue characteristics
themselves, it is possible to obtain a workable relationship
between acoustic measurements and lung pathology or the pathology
or condition of other biological tissues.
[0076] FIGS. 3A-3D are graphs of some indicators or parameters
involved with active auscultation. The information in FIGS. 3A and
3B is measured, whereas the information in FIGS. 3C and 3D is
derived. FIG. 3A is an exemplary chart of amplitude as a function
of frequency. FIG. 3C is an exemplary chart of attenuation as a
function of frequency, which is derived from frequency and
amplitude data such as that shown in FIG. 3A. FIG. 3B is an
exemplary chart of phase as a function of frequency, and FIG. 3D is
a chart of velocity as a function of frequency, which is derived
from the frequency and phase information such as that shown
[0077] Active auscultation can be used to assist in diagnosing lung
disease (also referred to herein as a condition) wherein again, a
sound is introduced to the thorax such that it travels from one
side of the thorax, through the lung, to another side of the
thorax. The sound velocity and preferably attenuation which is
measured is then compared with that of a normal, healthy lung.
Since lung disease often manifests in reduced lung volume, a
comparison can be used, again, to provide an indication as to
whether a subject's lung exhibits a propensity for lung disease.
Common lung diseases may include emphysema, asthma, regional
collapse (atelectasis), interstitial oedema and both focal lung
disease (e.g. tumour) and global lung disease (e.g. emphysema).
Passive Auscultation
[0078] The term "passive auscultation" generally refers to
monitoring an/or receiving at least one naturally occurring sound
from the body or a portion thereof for use in the diagnosis and/or
treatment of the body or a portion thereof. In relation to the
lungs, passive auscultation generally includes monitoring naturally
occurring breath sounds. This is in contrast to the active
auscultation described above where a signal is injected and then
received. The breath sounds are monitored and processed with system
100's receiving unit 116 and control unit 110.
[0079] Passive auscultation provides very important information
that a physician can use in understanding the condition of a
patient's lungs. Passive auscultation is well accepted by
physicians and provides valuable information that cannot be had via
active auscultation techniques. However, because it has been
difficult to quantify and lends itself more to subjective rather
than objective analysis, more attention has recently been focused
on the technique of active auscultation. This however, leaves out
an important piece of the puzzle that a physician may use to arrive
at a fully informed conclusion. System 100 utilizes both active and
passive auscultation data.
[0080] FIG. 4 is a chart of breath sounds. The acoustic
characteristics of these sounds are recognized by control unit 110.
Therefore, by monitoring these breath sounds during passive
auscultation, the system will have indications of the associated
condition of the lung. An adventitious sound from the passive
auscultation, when combined with an indicator from the active
auscultation may point the physician to a more narrow range of
possibilities and lead to a more accurate diagnosis. Likewise, a
normal sound heard during passive auscultation may lead to a
different range of possibilities of lung condition than would be
foreseen from an adventitious sound or from active auscultation
alone.
[0081] Adventitious lung sounds may be classified into three major
categories: crackles (or rales), which are discontinuous (i.e.,
interrupted) sounds, and wheezes and rhonchi, which are continuous.
Crackles may be further classified depending on their frequency,
characteristics and amplitude. Wheezes may be similarly classified.
An experienced and knowledgeable health care professional may also
be able to diagnose certain pulmonary diseases, such as pneumonia,
asthma, etc., simply by detecting, identifying and noting the
location of particular adventitious lung sounds. This can be done
with the aid of system 100 of the present invention.
[0082] Wheezing is a commonly used acoustical term in respiratory
medicine. Wheezes are well understood to be an indicator of airway
obstruction in infants, as a parameter to gauge the severity of
asthma, or as a classifier in epidemiologic surveys, to name just a
few examples. As can be seen in the chart of FIG. 4, a wheeze has
the acoustic characteristic of a sinusoid with a range of about 100
to above 1000 Hz and a duration typically, but not necessarily,
longer than 80 milliseconds. Generally speaking, a wheeze is
indicative of an airway obstruction or flow limitation. Wheezes are
"continuous" since their duration is much longer than that of
"discontinuous" crackles. In most cases they do not extend more
than 250 ms, but they will typically be longer than 80 to 100 ms.
Their frequency range extends from less than 100 Hz to more than 1
kHz, and higher frequencies may be measured inside the airways.
[0083] Spontaneous wheeze is often present during inspiration in
adults and children with asthma, but is not present in healthy
subjects. Regional flow limitation during inspiration is a
possibility but has not been proven. The sound of wheezing is
easily recognized since it stands out from the noise of normal lung
sounds. A wheeze of at least moderate intensity is characterized by
sharp peaks in the power spectrum of respiratory sounds. Detection
is possible with control unit 110 by comparing these spectral peaks
to the average lung sound amplitude.
[0084] Crackles are another type of easily recognizable
adventitious lung sounds. Crackles have been characterized as
"miniature explosions," and are heard much more frequently during
inspiration than during expiration. Crackles are sometime
categorized as either fine or coarse crackles. As can be seen in
the chart of FIG. 4 crackles have the acoustic characteristic of a
rapidly dampened wave deflection with a duration typically but not
necessarily less than 20 milliseconds. Generally speaking, a
crackle may indicate an airway closure or some type of airway
secretion.
[0085] It is now generally accepted that fine and coarse crackles
are associated with different conditions and so have diagnostic
importance. Crackles in patients with fibrotic lung diseases are
generally shorter in duration and period than the coarse crackles
of patients with pneumonia. The most commonly used indices are the
time duration of the initial deflection and the first two cycles of
the waveform.
[0086] Although crackle features and characteristics can be
associated with certain diseases, only certain associations are
currently well correlated enough to have clinical utility at the
moment. This is expected to change over time. Those with
established clinical utility appear to be: the presence or absence
of crackles to distinguish pulmonary fibrosis (crackles usually
prominent) from sarcoidosis (crackles usually minimal or absent);
fine, late inspiratory crackles indicating fibrotic lung disease
and early, coarse crackles indicating obstructive lung disease;
crackles as an early (perhaps first) sign of asbestosis, and
crackles indicating heart failure.
[0087] A rhonchus is similar to a crackle and is acoustically
characterized by a series of rapidly dampened sinusoids typically
from less than 300 Hz in frequency and with a duration greater than
100 milliseconds. A rhoncus is generally indicative a secretions or
abnormal airway collapsibility.
[0088] For more information on these conditions please refer to
"Respiratory Sounds Advances Beyond the Stethoscope" by Pasterkamp
et al., Am. J. Respir. Crit. Care Med., Volume 156, Number 3,
September 1997, 974-987, which is hereby incorporated by this
reference in its entirety.
[0089] Note that these characteristics and ranges obviously vary
from patient to patient and are only given as example ranges, and
should not be interpreted as limitations to the invention in any
way.
[0090] FIGS. 5A-5C are a few exemplary charts of some indicators or
variables of passive auscultation. FIG. 5A shows amplitude of the
monitored breath sounds as a function of time. A Fast Fourier
Transform or FFT extracts such information as seen in FIGS. 5B and
5C. FIG. 5B shows amplitude as a function of time, and FIG. 5C
shows phase as a function of frequency.
[0091] The present invention utilizes indications from both active
and passive auscultation to arrive at a more informed conclusion.
By combining or "multiplexing" lung condition indicators gathered
from both of these different techniques, a more narrow set of
indicators of lung condition is available. In the hands of a
trained physician or other professional, this will increase the
probability of more accurate diagnoses. System 100, and in
particular, control unit 110 processes the signal received from
monitored breath sounds of passive auscultation as well as the
signal injected and then received from active auscultation. The
system performs multivariable analysis using variables obtained
from both the passive and active measurements. Such analysis is
well understood by those of skill in the art. The number of
variables or condition indicators that can be utilized in such an
analysis is very large. Depicting even a three dimensional analysis
on (2 dimensional) paper is complex. The conditions, condition
indicators, and characteristics that can be determined with the
present invention include those earlier described and the
following: early stage emphysema; late stage emphysema; chronic
bronchitis; asthma; tissue micro-structure; alveolar dimensions;
fenestrae size; airway mucus loading; and lung viscosity. This is
not meant to be an exhaustive list and it should be understood that
with further study that various combinations of measured and
derived parameters will likely be associated with different
conditions, condition indicators, and characteristics of the
lung.
[0092] A few of the variables or indicators that have been
previously described above are illustrated in FIGS. 6-9.
[0093] FIG. 6 illustrates velocity as a function of frequency,
which is determined from active auscultation. Three curves and the
associated error bars are shown illustrating the regions where
these variables would indicate the lung condition as either normal
lung function, mild emphysema, or severe emphysema. As was
discussed at length previously, the higher the velocity of the
injected signal (along the frequency spectrum) the greater the
indication of emphysema.
[0094] FIG. 7 illustrates attenuation as a function of frequency.
The attenuation-frequency curve provides an indication of one or
more lung conditions. FIG. 7 illustrates attenuation at a different
range of frequencies (f.sub.3 and f.sub.4) than in FIG. 6, but it
should be understood that attenuation may also provide useful
information when determined in the same range as velocity shown in
FIG. 6. Different attenuation curves may indicate a different
disease state. For example, the solid curve is shown to be
indicative of "disease state b" whereas the dotted curve is shown
to be indicative of "disease state a."
[0095] FIG. 8 is a chart illustrating conditions associated with
crackle amplitude and wheeze amplitude. There exists a region where
both the wheeze amplitude and crackle amplitude are low, therefore
indicating a healthy lung. This is signified by "normal region"
810. A high wheeze amplitude may indicate an airway obstruction or
flow limitation, as is represented by region 810B, the region above
the horizontal dotted line. Also, a relatively high crackle
amplitude is indicative of an airway closure or secretions, as is
represented by region 810C, the region to the right of the vertical
dotted line.
[0096] As mentioned previously a large number of variables can be
combined in a multi-variable or multi-dimensional analysis. This
may include tens or hundreds of variables, but all that can be
represented on paper are two or three variables. FIG. 9 is an
example of a three dimensional analysis although higher dimension
analysis is preferable. Different lung states may be indicated
according to the levels of the various indicators. In this case,
the indicators include wheeze amplitude, velocity, and attenuation.
The attenuation is a ratio of the attenuation at different
frequencies, for example the ratio of attenuation at frequency 4
versus frequency 3.
[0097] A relatively low wheeze amplitude, velocity, and attenuation
ratio are indicative of normal lung function, as represented by
normal region 910A. A high velocity with little wheeze amplitude
and moderate attenuation ratio may together indicate "disease state
c." Similarly, a moderate wheeze amplitude together with a
relatively high velocity and attenuation ratio may point to
"disease state d."
[0098] While active auscultation provides excellent hard data, a
physician or other professional benefits from having as much data
as possible in any analysis. Combining information from both active
and passive auscultation, as is done in the present invention, is
far more beneficial than a conclusion based upon data from only one
or the other methods. This is especially true when a microprocessor
driven control system aids in the analysis as is the case with the
present invention.
[0099] The present invention is described herein with reference to
the accompanying examples and figures. It is to be understood that
the description is illustrative only and should not be taken to be
limiting in any way, or as a restriction on the generality of
applications for the invention previously described. Although the
various aspects and features of the present invention have been
described with respect to various embodiments and specific examples
herein, it will be understood that the invention is entitled to
protection within the full scope of the appended claims.
* * * * *