WO2000027287A1 - Non-invasive acoustic detection of coronary artery disease - Google Patents

Abstract

Multi-channel acoustic sensor array's configured to fit standardized acoustic windows or to operate within the bounds of a patient's acoustic window are described. One method employs SNR based weighting factors to discard or include specific sensors in the diagnostic evaluation (to include or exclude the sensor from the acoustic window based on its SNR weight). Medical application of acoustic array designs and geometries and/or apertures configured to accommodate patient acoustic windows and/or merged acoustic windows are exemplified.

Description

NON-INVASIVE ACOUSΗC DETECTION OF CORONARY ARTERY DISEASE

RELATED APPLICATIONS

This application claims priority from U.S. Patent Application No. 09/188,434 filed 9 November 1998 and provisional U.S. Patent Application No. 60/107,616 filed 9 November 1998. The contents of these applications are incorporated by reference as if recited in full herein.

FIELD OF THE INVENTION

This invention relates to arrays of acoustic sensors that facilitate the non- invasive detection of coronary artery disease (CAD).

BACKGROUND OF THE INVENTION

PCT Application No. PCT/US97/20186 filed November 10, 1997 ("the 20186 application") entitled "Non-Invasive Turbulent Blood Flow Imaging System" describes an invention for the non-invasive in vivo detection and localization of abnormal blood flow. The contents of this application are hereby incorporated by reference herein. Embodiments of that invention display the spatial distribution of phase coherence in the shear eave component of blood flow signals generated by an acoustic sensor array. An essentially uniform display indicates normal blood flow. A non-uniform display may indicate the presence of an occlusion and the presence or extent of abnormal, turbulent blood flow. Poor correlation of signals from the array sensors may adversely affect the display uniformity.

Acoustic sensor arrays are conventionally positioned above a measurement area defined as the hairless human chest skin located vertically between the sternum and a parallel line passing through the left nipple and horizontally 10 cm above and 6 cm below the left and right nipples.

A prior art acoustic sensor array comprising eight equally spaced sensors in two concentric circles having prime numbers of sensors in each circle and a ninth sensor at the common center ofthe concentric circle is illustrated by Figure 6 ofthe 20186 application.

To reach sensors in a conventionally positioned prior art array as described in the 20186 application, sound waves must travel either directly through lung tissue or first to the body surface and then laterally with consequent attenuation of correlation. A study ofthe correlation by that array of patient data signals generated by the quiet interval revealed that only four or five ofthe nine sensors are well correlated.

It is known that a notch (a cardiac notch) in the human left lung allows the heart to be in contact with the chest wall. Well-correlated blood flow signals may be generated by acoustic sensors positioned on a human chest in a small area (the region identified as the acoustic window) located above the cardiac notch.

However, there remains a need to be able to provide improved ways to identify the acoustic window and/or improve signal correlation for improved sensor operation and clinical applications.

DEFINITIONS

Acoustic Window— An area above the notch in the human left lung which allows the heart to be in contact with the chest wall. Well-correlated acoustic blood flow signals of good quality may be generated by a sensor array positioned on a patient's chest within or substantially within the perimeter of an acoustic window.

Sensor or Accelerometer— Any current or voltage mode device which generates an electric signal from displacement or a derivative thereof upon detection of a sound wave.

Sensor Array— A pattern or spaced arrangement of a plurality of sensors on or to be placed on the body surface of a patient. For the purposes of this invention, the sensor array is an acoustic multi-channel sensor array with at least four sensors and a corresponding four sensor channels. Sensor Array Aperture— The space or area within the perimeter of an array where heart or blood flow sounds are detected by a sensor(s) positioned therein.

Sensor Array Geometry— The shape ofthe perimeter of a multi-channel sensor array or array arrangement on a subject's chest.

Channel-The path from a sensor to a receiver followed by a signal from the sensor by which the signal is generated. A sensor array includes multiple sensors and multiple channels.

Weight— For the purposes of this invention a weight is a constant applied to the SNR on single sensor channel as indicative of its relative importance among all involved channels. An algorithm for computing the weights, and preferably the optimal weights for each channel scaled to the estimated SNR thereon is described in the 20186 application, and herein at Appendix A. The algorithm operationally corresponds to and or depends on having the same sensor location in all measurements for a particular person.

SUMMARY OF THE INVENTION

The present invention employs methods for determining an acoustic window suitable for passive-acoustic coronary artery disease evaluation. The bounds ofthe acoustic window can be approximated by ultrasonic probe means or signal to noise ratio (SNR) evaluation means. Array sensors can then be configured such that the sensors are arranged in positions which correspond to channels achieving the highest apparent SNR or which correspond to a statistically determined acoustic window site as described in this application. The present invention thus provides multi-channel acoustic sensor arrays which are configured to exploit the size/configuration ofthe acoustic window to thereby allow for improved acoustic or CAD clinical diagnostic applications.

Pursuant to one embodiment ofthe invention, an acoustic window may be defined by ultrasonic probe means. The invention includes sensor arrays having an aperture locatable within or substantially within the bounds of an acoustic window when the array is positioned on the chest of a person. In a preferred embodiment, the invention includes the identification of an acoustic window comprising the merged acoustic window sub-areas corresponding to two or more intercostal spaces (ICS's), and array designs to accommodate such acoustic windows.

Another method includes the steps of (a) positioning a multi-channel acoustic sensor array (preferably having at least four and more preferably about 9-45 sensors) onto the chest of a subject; (b) calculating a weighted value for each ofthe sensor channels in the multi-channel sensor array; (c) determining the location of each sensor channel in the array; (d) identifying the sensor channels which meet predetermined test criteria; and (e) defining a perimeter which substantially extends about and encloses therewithin the sensor channels identified in step (d), thereby defining an acoustic window suitable for acoustic listening diagnostic procedures.

In a preferred embodiment the calculating step is performed by assigning signal to noise ratio (SNR) based weighted values to each ofthe sensor channels and the predetermined test criteria includes identifying the sensors exhibiting the three highest calculated weighted values or identifying at least three sensors exhibiting one or more high weight values. The acoustic window can be used to define one or more standard optimum sensor array geometry and sizes.

This method involves the discovery that an acoustic window can be visualized by grayscale or equivalent mapping of optimal weights scaled to the estimated SNR on each of a plurality of channels of a multichannel acoustic sensor array to the nominal location of each sensor. The grayscale maps identify channels that achieve the highest SNR because the optimal weights represent a measure ofthe relative SNR distributed at each ofthe nominal sensor locations.

In operation, as shown in Figure 24, the bounds of an acoustic window are visualized or defined by a perimeter (shown in dotted line) that encloses three or more channels that exhibit the highest relative SNR as measured by the optimal weights. The acoustic window is used to bound the aperture of an acoustic sensor array. This acoustic window identification increases or enhances the probability of acquiring the highest possible SNR on the largest percentage of sensors in the array.

Advantageously, the present invention allows the acoustic array and the array sensors to be located to fit within acoustic windows for improved numbers of well-correlated sensors. The present invention identifies acoustic window sizes across a population can allow standardized acoustic array sizes which statistically correspond to particular population segments or one which operably across a large population.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 illustrates an ultrasonic probe acoustic window characterization method that provides a template for the positioning of sensors on a person's chest. All acoustic window data illustrated by Figures 2 through 8 and 10 through 12 was obtained by the Figure 1 method.

Figure 2 is a plot in polar format ofthe acoustic window size data obtained from 22 male and 7 female subjects (29 subjects).

Figure 3 is a histogram ofthe window areas ofthe same 29 subjects from which the Figure 2 data was obtained.

Figure 4 shows acoustic window size in polar format. Maximum, minimum and average window size for all ofthe same 29 subjects is depicted.

Figure 5 illustrates in Cartesian coordinates variations ofthe ICS ultrasonic probe data points from the left (□) and right (0) ICS. The statistical averages for ICS's 1 to 5 are shown in solid lines. A perimeter connecting the ends ofthe solid lines is a visualization ofthe average geometry ofthe six intercostal spaces.

Figure 6 is a histogram that indicates ICS nearest to the centroids ofthe average window area (see Figure 4).

Figure 7 is a histogram indicating the distribution ofthe perpendicular distances from the centroid ofthe average window area (see Figure 4) to the nearest ICS. Figure 8 is a histogram illustrating the distribution ofthe distance from the left side ofthe nearest ICS to the projection ofthe centroid ofthe average window area (see Figure 4).

Figure 9 depicts a prior art nine sensor array based on seismic accelerometers commercially available from Wilcoxon Research, 21 Firstfield Road, Gaithersburg, Maryland 20878. The array comprises eight equally spaced sensors in two concentric circles having prime numbers of sensors in each circle and a ninth sensor in the common center ofthe concentric circles.

Figure 10 depicts a 13 element array positioned over an acoustic window of average size (solid line, see Figure 4). An acoustic window of maximum area is also shown (broken line).

Figure 11 illustrates a 57 element small PVDF sensor array based on averaging of ICS data points. The array comprises five linear subarrays positioned above intercostal spaces 2 to 6.

Figure 12 illustrates a 32 element array of relatively large PVDF sensors based on averaging of ICS data points. The array comprises five linear subarrays positioned above intercostal spaces 2 to 6.

Figure 13 illustrates a beam pattern in x for delay-and-sum (DS) (higher in value) and MVDR (lower in value) beamformers using prior art (HA) (dashed lines), bowling pin (BO) (dashed dotted lines), small PDVF (SP) (solid lines) and large PDVF (LP) (dotted lines) arrays.

Figure 14 illustrates a beam pattern in y (dB) for delay-and-sum (DS) (broken lines) and MVDR (solid lines) beamformers using prior art (HA), bowling pin (BO), small PDVF (SP) and large PDVF (LP) arrays.

Figure 15 illustrates a beam pattern in z (dB) for delay-and-sum (DS) (broken lines) and MVDR (solid lines) beamformers using Harris (HA), bowling pin (BO), small PDVF (SP) and large PDVF (LP) arrays.

Figure 16 is a proposed array based on the medical Wilcoxon accelerometer. Only 12 elements are used due to the limitation ofthe data collection system. The broken line indicates the perimeter of an acoustic window. Figure 17 illustrates a 45-element PVDF sensor array comprising five nine element linear subarrays positioned above intercostal spaces 2 to 6.

Large and small acoustic window perimeters with centroids near the fourth and fifth ICS's are shown.

Figure 18 is an image or display that depicts optimal weights scaled to the estimated channel SNR grayscale mapped to nine sensors in a conventional nine- sensor array as observed with a human patient. The sensors and corresponding channel are numbered 1 to 9. Maximum channel weights (the darker grayscale images) are shown for the sensors in the center and to the patient's left.

Figure 19 is an image or display that depicts optimal weights scaled to the estimated channel SNR grayscale mapped to a nine-sensor array having the pattern indicated as observed with a human patient. The channel numbers and associated channel optimal weights are superimposed on the grayscale images, e.g., 1/0.4.

Figure 20 is a histogram graph indicating the distribution of channels with maximum (highest relative SNR) mean optimal weight. The graph represents the maximum mean optimal weight for each channel across a population of patients (taken from about 100 intervention and non-intervention or not significantly diseased (NSD) patients) using the Figure 18 conventional nine-sensor array configuration.

Figure 21 is a histogram graph including indicating the distribution of channels with maximum mean (highest relative SNR) optimal weights. This graph represents the weights and channel distribution for intervention patients only (using the Figure 18 conventional nine-sensor array configuration). The results are similar to that illustrated by Figure 20 for all patients.

Figure 22 is a histogram depicting the distribution of channels with maximum channel (highest relative SNR) mean optimal weights. This graph represents only the non-significant disease (NSD) patients (using the Figure 18 prior art nine-sensor array configuration).

Figure 23 is a histogram depicting the distribution of channels with maximum mean optimal weight. This graph represents the NSD patients and uses the nine- sensor array configuration of Figure 19.

Figure 24 is a duplicate of Figure 18 upon which a broken line has been interposed to provide visualization of an acoustic window including sensors 4 to 9 assigned the highest optimal weights according to the present invention.

Figure 25 is a block diagram of method steps used to determine an acoustic window according to the present invention.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

The present invention now will be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments ofthe invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope ofthe invention to those skilled in the art. Like numbers refer to like elements throughout. In the figures, certain layers, regions, or components may be exaggerated or enlarged for clarity.

The invention generally comprises the identification of an acoustic window and the design of arrays having geometry sized to fit within or substantially within and thus accommodate the perimeter ofthe window. The invention may include an ultrasound derived average acoustic window and consolidated or merged window subareas and array geometry sized accordingly. The invention also comprises means for identifying or visualizing an acoustic window by mapping (relationally determining the position of) the nominal sensor locations ofthe weights associated with each sensor channel, and more preferably to optimal weights scaled to the estimated SNR on each of at least four channels of an acoustic sensor array.

Ultrasonic Probe Determination of The

Size and Location ofthe Acoustic Window—

Design of Template for Sensor Positioning

The acoustic window may include one or a combination ofthe small areas (intercostal space window areas) ofthe patient's chest surface directly above the intercostal spaces one through six. Determination ofthe size of an acoustic window may be accomplished by steps (i) to (v).

(i) With the patient supine, i.e., lying on his back or side, draw a series of dots along the left sternal border at the beginning of each intercostal space (ICS) for spaces one through six.

(ii) Place an ultrasound probe at the left sternal border ofthe first intercostal space (ICS). Then move the probe along the intercostal space until the lung tissue is encountered. Place a dot on the chest to mark where the lung tissue begins.

(iii) Repeat step (ii) for intercostal spaces two through six.

(iv) Wipe the ultrasound gel off the chest, and draw a line following each intercostal space, connecting the two previously drawn dots. The lines should be similar to Figure 1.

(v) After the chest has been marked as above, place a sheet of tracing paper on the chest, and transfer the markings onto the paper to provide a template for positioning of sensors.

Average or "generic/standardized" templates may be statistically prepared from (average) data determined in the same way from a plurality of persons.

A similar procedure may be used to determine an acoustic window of a person lying slanted on a bed.

Table 1 lists the window areas (in cm²) for the three window sizes in two bed positions. For the maximum and average window cases, lying slanted on the bed produces a slightly larger (5% and 16%, respectively) window size than lying flat on the bed. In the case ofthe minimum window, lying flat does produce a significant 37% larger window area than lying slanted on the bed. Table 1 Differences in the Flat and Slanted Bed Positions

Based on these results, the flat position is more advantageous since it does not significantly reduce the acoustics window for subjects with large to medium window sizes and at the same time significantly opens up the smaller acoustic window sizes.

Another method for identifying an acoustic window entails examination of which sensor channels receive the highest signal to noise ratio (SNR) as measured by the optimal weights for summing channels will be described further below under the section entitled Acoustic Window Identification.

Statistical Analysis of Acoustic Window Data Acoustic window size data, collected pursuant to the described ultrasonic probe methodology, was obtained from 22 male and 7 female subjects. There are two types of data.

1. Measurements based on a Cartesian coordinate with X axis on the 6^th intercostal space (ICS) and Y axis along the left end ofthe ICS.

2. Data estimate in polar coordinate centered at the centroid ofthe acoustic window mass. This data is derived from measuring the distance from the centroid to the edge ofthe window at 30 degree angle increments. There are a total of twelve data points per subject. The Distribution of Polar Data The distribution ofthe window size data in polar format was tested for normality using the normal probability plot from the MATLAB Statistics Toolbox.

The purpose of a normal probability plot is to graphically assess whether the data could come from a normal distribution. If the data are normal, the plot will be linear.

Other distribution types will introduce curvature in the plot. As shown in Figure 2, the data points are virtually in a straight line, indicating that the polar coordinate data is Gaussian.

When the area for each ofthe 22 male windows was computed using AutoCad software, the ratio ofthe maximum to the minimum area was found to be 15. Figure

3 illustrates the histogram ofthe window areas.

Figure 4 shows the acoustics window size in polar format. In this figure, the outside perimeter is for maximum, the inside perimeter is for the minimum and the intermediate perimeter is for the average across all subjects.

Correlation Coefficients Between Acoustics Window Areas and Demographic Data

Computed correlation coefficients between the window area and the subject demographics data are shown in Table 2. This analysis was carried out based on data broken down by male (22 subjects), female (7 subjects) and a combination of both sexes. In general, there exists no strong correlation between the window area and demographics data, with the exception of strong negative correlation of 0.84 between the acoustics window area and the anterior/posterior (AP) diameter ofthe female subjects and a strong negative correlation between the acoustics window area and the sternum length in both male and female subjects. Table 2

Variations ofthe ICS End Points in Cartesian Coordinate

Examination ofthe data in Cartesian coordinate reveals the absence of common single reference point such as the centroid in the polar data case. The X- Y data was collected relative to the six intercostal spaces and was measured as left and right ICS. The only single common reference was made when the six ICS's were aligned on the x-axis so that the other spaces can be seen relative to this reference space.

As seen from Figure 5, the XY coordinates ofthe left in (□) and right in (0) ICS are quite different across the subjects. The statistical averages of the left and right ICS are also shown. When connected, a visualization ofthe average geometry ofthe six intercostal spaces is provided. The Distribution of Window Centroids Over ICS The x-y coordinates ofthe acoustics window centroids were measured and correlated with the lines defined by the left and right ICS. The purpose of this correlation is to determine which ofthe six spaces the window centroid is near to and then to ascertain the best space(s) for location ofthe array.

A histogram ofthe ICS to which the centroids ofthe window areas are nearest is plotted in Figure 6. The result indicates that the fourth and fifth ICS are good candidates for positioning the array center, with the fourth ICS being more frequent than the fifth ICS. In practice, it is appropriate to consider both of these two ICS equally and pick one based on the best knowledge of which ICS has the best heartbeat sound.

The distribution of (a) the perpendicular distances from the centroid to the nearest ICS and (b) the distances from the left side ofthe nearest ICS to the projection ofthe centroid are histogrammed in Figures 7 and 8. These results provide guidelines as to the approximate location ofthe array center relative to the nearest intercostal space.

Array Design Based on Acoustics Window Data Factors and Constraints in Array Geometry Design

Design constraints considered or imposed on array geometry may include:

1. Limitation on the array aperture by the size ofthe acoustic window which varies from person to person.

2. Sensor size which limits the number of sensor elements that can be put in the array aperture. For example, the medical Wilcoxon sensor diameter is about one cm. In the case of PVDF sensors, the prefabricated thin film strip size can dictate the number of sensors which can be placed within the array aperture.

3. The anti-aliasing requirement ofthe array design at different operating frequencies. In principle, the inter-element spacing ofthe sensors is required to be less than half a wavelength at the highest operating frequency to avoid spatial aliasing in the plane wave case. This requirement is relaxed in the near field where source location is the objective. The use of irregularity in array geometry may also alleviate the aliasing problem when there is an inter- element spacing of more than half wavelength.

Array Geometries

Use ofthe acoustics window configuration in array design based on the medical Wilcoxon accelerometer and the large and small PVDF sensors resulted in the three exemplary arrays depicted by Figures 10, 11 and 12.

The 13-element array of Figure 10 was based on the average window size described with reference to Figure 4. Using the actual dimensions ofthe medical Wilcoxon accelerometer on graph paper, each accelerometer was placed on straight lines starting from the center and populating the perimeter until space is occupied. A total of 13 elements that were fitted into this average window size. Clinical data indicates that 13 elements may not give optimum array gain especially when element signal-to-noise ratio of turbulent flow is low.

More elements per unit area are possible with PVDF technology. Because PVDF film is available in linear strips of multiple sensory each strip (such as 6, 9 and 16 elements per unit), each strip can be put on the intercostal space to maximize signal reception. These factors motivate exemplary array geometries illustrated in Figures 11 and 12.

In these two arrangements, five lines of PVDF film strip are placed along ICS's 2 to 6 at approximately the length ofthe average ICS as described with reference to Figure 5. The placement of these PVDF film strips as shown in Figures 11 and 12 are for illustration only and not necessarily the exact position and direction ofthe film strips. Also, because ofthe inherent variations in human anatomy, the actual placement ofthe PVDF sensor strips is expected to be different from person to person, in view ofthe effect ofthe ribs as a factor in signal reception.

It is noted that 57 elements for the small PVDF and 32 elements for the large PVDF sensors were used in this array design. Array Performance

The performance ofthe four sensor arrays depicted by Figures 9, 10, 11 and 12 is presented in terms of beam width and array gain by Figures 13, 14 and 15. The beam pattern plots are for frequency at 250 Hz using Verberg propagation model and 10 dB element SNR. These figures show the beam patterns in x, y and z direction for a source 3 cm directly below the array center. The beam pattern for the conventional delay and sum (DS) beamformer is shown in dashed line, and the beam pattern for the MVDR beamformer is in solid line.

The figures show that for a conventional beamformer, the array gain is proportional to the number of elements. The effect ofthe number of elements on the array beamwidth is much more visible for the MVDR beamformer than for the DS beamformer. Also notable is the lack of array aperture in the z direction, as illustrated by the large beam width shown in Figure 15.

It is known that the MVDR beamformer provides an estimate ofthe signal power at the signal direction as can clearly be seen from Figures 13, 14 and 15. At the source location, the output ofthe MVDR beamformer is 10 dB regardless ofthe number of element in the array. The effect of an increase in the number of elements is a narrower beam width, which is consistent with data showing that the beam width of an MVDR beamformer is inversely proportional to the number of elements (and the element SNR).

Array Geometries

Increasing the number of elements is constrained by the acoustic window size and the physical dimensions ofthe individual sensor as noted above.

The instant invention recognizes that from the performance ofthe array designs of Figures 10 to 14 that the use ofthe 4^th and 5^th intercostal spaces for centering purpose has merit in the array design process. For both designs, the acoustics window is the union ofthe two window areas for the 4^th and 5^th ICS's. These two windows are the average ofthe XY data obtained from the acoustics window study. The merging of the two windows increases the area available for the array aperture which is an advantage to array performance. A Proposed Wilcoxon Accelerometer Array Based on the Dominant ICS Areas

A design for a Wilcoxon commercial accelerometer array is shown in

Figure 16. In this design, there are a total of 16 elements that will substantially fit the composite window area. The composite window area consists ofthe two averaged windows with centroids near the 4^th and 5^th ICS's.

In the current data collection system, only 12 elements are used. The missing elements are chosen such that the resulting array is as irregular as possible, and preferably configured with at least one sensor pair very close to each other to prevent spatial aliasing.

A PVDF Sensor Array For the same composite acoustics window, a 45-element PVDF array is shown in Figure 17. This array essentially consists of 5 rows of 9-element large PVDF linear array arranged in such a way that conforms to the human chest curvature and, if possible, lies within the lower ICS to adapt to the patient anatomy. One reason for a 5 by 9 linear PVDF array is in the manufacturing and logistics ofthe thin film technology.

It may not be possible to use all 45 elements for beamforming, since some ofthe array elements may fall out the acoustics window and thus will not be able to receive the heart sound (or be inhibited by lung tissue and the like). An acceptable data collection scheme includes estimation ofthe signal- to-noise ratio at each element, and weighting or eliminating the sensors that receive the noisiest signal. The use of this weighting technique enables the array to adapt to the differences in acoustic window size that are embodied in human anatomy. This weighting technique can be used in addition to the window identification and array design parameters discussed above, or can be used independently as will be discussed further below.

Of course, the invention is not limited to the particular array configurations described herein as other multi-channel sensor arrays can also be used. For example, arrangements of other array geometries, preferably employing a plurality of strips which are substantially linear as shown in Figure 1, each strip having a plurality of sensors and corresponding sensor channels included thereon. See also, sensor arrays described in co-pending and co-assigned US Patent Application Serial No. 09/136,933, entitled "Thin Film Piezoelectric Polymer Sensor" and US provisional application Serial No. 60/132,041 filed April 30, 1999, entitled "Low Profile Sensors". The contents ofthe above-identified applications are hereby incorporated by reference as if recited in full herein.

Acoustic Window Identification

One preferred embodiment for identifying or visualizing an acoustic window maps (relationally determines the position of) the nominal sensor locations associated with the weights assigned to each sensor channel, and more preferably to optimal weights scaled to the estimated SNR on each of at least four channels of a multichannel acoustic sensor array.

As shown in Figure 24, the bounds ofthe acoustic window are visualized by a perimeter which encloses at least the three nominal sensor locations that correspond to the highest optimal channel weights. Any array that comprises four or more sensors, and therefore, four or more channels, may be used to practice the weighted method of invention. Preferably, the sensor arrays used according to the present invention are configured to have from about 9 to 45 channels and preferably from 9-45 corresponding sensors.

Figures 18 and 19 grayscale image maps show optimal weights as distributed at the nominal sensor locations for two different nine-sensor array configurations. Figure 18 illustrates the conventional nine-sensor array (concentric arrangement) of the 20186 application and Figure 19 illustrates a different nine-sensor array configuration. The channel number and weight are identified according to its sensor location.

The present invention compares the weighted channels to identify the particular channels corresponding to sensors within a multi-channel sensor array which meet a predetermined or relative threshold value. Preferably, the sensor locations ofthe channels with the three largest weights (or three sensors with one or two ofthe largest calculated weights) are used to define the acoustic window perimeter. The acoustic window perimeter is drawn to extend and include the sensors corresponding to the three largest weighted channels and may also include sensors corresponding to channels with lesser weighted values, such as shown in Figure 24. As noted the predetermined threshold value(s) can be an absolute number, i.e., such as those at above about a 0.7 optimal value or a relative threshold such as a sliding scale which is set such that it includes at least three sensors or channels. Accordingly, more than three channels may have relatively high weights and the acoustic window will be drawn to include at least the sensors corresponding to those channels. For example, one channel may have a weighted value of about 0.8 while three channels may have a 0.7 weighted value. In this instance, sensors corresponding to all four channels will be preferably included within the acoustic window.

That is, referring again to Figure 24, a plurality of increased or relatively high weighted channels (channels 2, 3, 4, 7, 8, 9) are identified (having the top three weighted values of 0.9, 0.7, and 0.5). In order to include the sensors corresponding to these channels, the acoustic window perimeter extends from the sensor locations for channels 2 and 7 on the top to the sensor location for channel 3 on the bottom and over to the sensor locations for channels 8 and 9 and 4 to the patient right and the sensor location for channel 2 on the patient left.

Because the spatial variation ofthe weights is smooth across the arrays, it is appropriate to consider the channels with the highest weights as indicators ofthe sensor locations within the bounds of an acoustic window. The channels with the maximum mean (across beats) optimal weights are found and shown in Figures 20 through 23 and are drawn within the acoustic window by the broken line perimeter in Figure 24.

Figures 18 and 24 show an acoustic window that extends in the x- direction from about 4 cm to about 10 cm below the 2^nd intercostal space (ICS) and in the y-direction. Sensors to the patient's left are favored. Channels 3 and 4 appear to be close to the 6^th ICS where the window widens. In Figure 24, channel 4 is included in the window. The nine sensors ofthe sensor array of Figure 19 are placed in the intercostal spaces. Sensor 1 is placed in the 2^nd ICS at the left-sternal border, just as it is in the prior art array of Figures 18 and 24. The second row (sensors/channels 2 and 3) is placed in the 3^rd ICS and so on. The results for this Figure 19 configuration show an acoustic window that covers the area left of the 3^rd ICS and then follows the sternum covering the 4^th and 5^th intercostal spaces.

Combining these results for Figures 18 and 19 yields an acoustic window that starts at the left ofthe 3^rd ICS and then follows the sternum covering the 4^th through 6* spaces, widening to the right at the 6^th space. This is the window is visualized by the superposed broken lines in Figure 24.

In the nine-sensor array ofthe 20186 application (Figures 18, 22 and 24), two modes appear in the histograms: one at channels 7, 8, and 9 and a second at channels 2, 3, and 4. These modes are relatively unchanged when the analysis is done for interventional and non-significant disease patients. See Figures 20 and 21. For the Figure 19 array, modes appear at channels 3 and 4 and at channels 6 and 7. All of these patients were normal or non-significant diseased patients.

Because the relative (preferably optimal) SNR channel weights are independent ofthe types or sizes of sensors, the present invention may allow for acoustic window visualization without regard to sizes or types of sensors used. One embodiment ofthe invention provides acoustic window visualization by use of any piezoelectric film, e.g., polyvinylidene difluoride or PVDF sensors. Useful PVDF sensors are described in commonly assigned United States patent application Serial No. 09/136,933 filed August 20, 1998 and United States provisional application Serial No. 60/132,041 filed April 30, 1999. In one embodiment, at least four linear PVDF strips with six sensors each are configured to extend substantially proximate to the acoustic window during clinical acoustic-based CAD evaluation. Other embodiments ofthe invention include acoustic window visualization by use of commercial seismic or medical accelerometers available from Wilcoxon Research, 21 Firstfield Road, Gaithersburg, Maryland 20878 or piezoelectric film sensors available from MSI (formerly AMP Incorporated), 449 Eisenhower Boulevard, Harrisburg, Pennsylvania 17111-2302. The data reported in this application was obtained with arrays of commercial Wilcoxon accelerometers. Appendix A describes a suitable method for weighting sensor channels using SNR values of a multi-channel sensor array. Of course, alternate SNR evaluations can also be used.

In operation, the acoustic window of a person is preferably determined in conjunction with an acoustic-based non-invasive diagnostic evaluation, based on the following method steps. Alternatively, the acoustic window can be statistically defined by correlation of measurements for each array geometry of interest according to the following method steps across a population of patients. If the latter, the population based acoustic window determination can be used to define a preferred array geometry (one having improved SNR channels, by sizing and configuring the location ofthe sensors and the array geometry itself to fit substantially within the computationally- identified and correlated acoustic window). In addition, the methods ofthe present invention can be used to define additional correlation and a corresponding set of multi-sensor array configurations to allow for further customized clinical applications. For example, the acoustic window and preferred array configuration and size can be defined and classified corresponding to demographic or other representative features such as height, weight, chest size, and/or gender to further customize sensor array geometry (and providing appropriately sited/standardized array sets) particularly suitable within target patient groups for ease of selection at the clinical point of operation.

Referring to Figure 18, a multi-channel sensor array is positioned onto the chest of a subject undergoing evaluation (Block 100). A plurality of weighted-values are calculated, one for each ofthe sensor channels in the multi-channel sensor array (Block 110). The weights are associated with the SNR values of each signal for each channel. Preferably, the SNR-based weighted values are computationally scaled to an estimated SNR and mapped (electronically). In addition, the position or spatial location of each channel is identified (Block 120). Preferably, the position is identified relative to its location within a substantially fixed (known geometric) configuration multichannel sensor array (Block 121). The position can also be otherwise established, such as by identifying each sensor position relative to a spatial axis or grid system and/or relative to a particular ICS/chest location. However established, the relative location of one sensor to another sensor in the sensor array may be utilized to define an acoustic window with respect to other sensors in the array according to the present invention. It is also preferred that the sensor array be substantially consistently positioned on the chest within an estimated acoustic window across different subjects and/or on the same subject in subsequent procedures.

For each known geometrical relatively constant array configuration, the relative position of each sensor within that array can be electronically represented and identified by mapping the known spatial relationship between the location of each sensor to the others. Further, over a population of subjects, a statistical correlation ofthe size and shape ofthe acoustic window (based on number of high SNR weights/channels) as it relates to a particular array configuration can be used to size and configure the array to positively affect the diagnostic operation ofthe sensors. As such a set of differently sized array geometries can be provided to correspond thereto (to allow a clinician easy access to different sizes at the point of application). Alternatively, or in addition to the improved sizing ofthe array, the method can be used to define which channels are "active" during diagnostic listening corresponding to the high-weight channels or channels located within the acoustic window at the time ofthe procedure. (The acoustic window may be statistically consistent across a particular population, but accurate or repeatable positioning may vary procedure to procedure). Thus, this method, if performed contemporaneously, may provide additional improvements in diagnostic capabilities. In any event, it is preferred that the array comprises at least four sensors and a corresponding four channels.

Next, the sensor channel weights which meet predetermined criteria are identified (Block 130). This predetermined criteria can include one or both of absolute or relative criteria. For example, establishing a minimum threshold weight value and identifying which sensors have weighted values which meet or exceed the minimum threshold values (and/or identifying and subsequently excluding those that fail to meet minimum values) (Block 131). Alternatively, or in addition to the absolute criteria, relative or floating criteria can be employed (Block 133). For example, identifying the sensor(s) having the three largest weighted values. As another relative example, the method may identify the largest weighted value calculated and then count the number of sensors associated with channels exhibiting this value. If the number of sensors for channels having this value is less than three, then the next largest value is identified and the number of sensors associated with these channels having this value are counted, etc. This procedure can be repeated until a desired number ofthe sensors within the multi-sensor array are identified (preferably at least three sensors). Of course, combinations of absolute and relative test criteria can also be employed. Preferably, less than all ofthe sensors in the array are identified by the predetermined test criteria evaluation (i.e., at least three, but less than all ofthe sensor channels will typically correspond to the sensor channels determined to have high (and preferably optimal) weights).

It is also possible to establish the predetermined test criteria to identify any channel which should be excluded from consideration during diagnostic procedures based on its failure to meet certain minimum threshold criteria (increased signal interference or those channels exhibiting low weighted values) to thereby exclude sensors corresponding to channels which may be blocked by the presence of undesirable acoustic path interference (such as that associated with lung tissue) within the chest area. This method can be performed independent of or in addition to the increased or high-weight value method described above.

Based on the (high-weight) sensor channels identified by the predetermined test criteria evaluation step, a perimeter can be defined to extend about and enclose the sensors corresponding to channels meeting the predetermined criteria, thereby defining an acoustic window region on the chest of a subject (Block 140). That is, at least figuratively, a perimeter line can be drawn (electronically) about the sensor locations (which correspond to chest locations) which exhibit the high-weight values to define the bounds or outer limits of an area or region on the subject's chest corresponding to the acoustic window.

Preferably, the multi-channel sensor array has a perimeter and an associated aperture (the overall size ofthe array) and the array is configured such that the perimeter ofthe sensor array substantially conforms to (and/or extends beyond) the bounds of an acoustic window that starts at the left ofthe third intercostal space, follows the sternum covering the fourth through six intercostal spaces, and widens to the right at the sixth intercostal space of a person.

It will be understood that each ofthe method steps, block diagrams (or blocks in a flowchart illustration), and combinations of blocks in flowchart illustrations or blocks in block diagram figures), can be implemented by computer program instructions. These computer program instructions may be loaded onto a computer or other programmable data processing apparatus to produce a machine, such that the instructions which execute on the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowchart block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart block or blocks. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks and/or block diagrams.

Accordingly, the method steps, blocks ofthe block diagrams or in a flowchart illustration support combinations of means for performing the specified functions and program instruction means for performing the specified functions. It will also be understood that each block ofthe block diagram or flowchart illustrations, and combinations of blocks in the block diagrams or flowchart illustrations, can be implemented by special purpose hardware-based computer systems which perform the specified functions or steps, or combinations of special purpose hardware and computer instructions.

The foregoing is illustrative ofthe present invention and is not to be construed as limiting thereof. Although a few exemplary embodiments of this invention have been described, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. Accordingly, all such modifications are intended to be included within the scope of this invention as defined in the claims. In the claims, means-plus-function clause are intended to cover the structures described herein as performing the recited function and not only structural equivalents but also equivalent structures. Therefore, it is to be understood that the foregoing is illustrative ofthe present invention and is not to be construed as limited to the specific embodiments disclosed, and that modifications to the disclosed embodiments, as well as other embodiments, are intended to be included within the scope ofthe appended claims. The invention is defined by the following claims, with equivalents ofthe claims to be included therein. APPENDIXA

To determine the optimal weights for summing the .channels together let -^£ be the ratio of signal to noise. The signals, if time aligned, will sum coherently by the magnitude of the weights, while the noise will sum incoherently as the square- root of the sum of the squares of the noise i each channel times its respective weight . The equation becomes the ratio of these two relations. For the case of four channels:

ps_ - — & ^{+ W : S + W S + W S}*

where:

Si = signal on the ith channel,

Ni = noise on the ith channel (white and orthogonal to the noise and the signals on the other channels, and i = real valued weight for ith channel.

The maximum can be found by taking the partial derivatives of the signal to noise ratio with respect to the weights and setting them equal to zero. The four partial derivatives are shown below.

dR

- S

^ J + J + k*₃? + h* J?

_ ^1 + "A + ^₃ ^g ₃ + ^WΆ jfoff ² + [π₂N₂]² + [w₃N₂]² + [w.N,]² dR

dw_^ lwN_x]² + [w₂N₂Y + [w₃N₃]² + [w,N,

[w_xS_x + w₂S₂ + w₃S₃ + w,s y₂N r; jj -N ² + [w₂N₂]² + [w₃N₂Y + [w_AN_ΛY]

dR

9w,

+ [w₂N₂Y + [w₃N₃]² + [w,N,γ

dR

dw_A w,N_xY + [w₂N₂Y + [w₃N₃]² + [w.N.YJ

Multiplying the first term in these equations by one, written as-

and setting them equal to zero and simplifying yields:

For these equations to go to zero, the numerators must go to zero, which yields the following relationships:

If a priori information exists on the S and N of each channel then let _x = 1, then the other weights can be found. If the S and N for each channel must be found a pos ori, then the correlation coefficients can used to generate estimates of S and N for each channel J The matrix of peak correlation values between channels must be found to do the time alignment . By the following method, the column or row of the peak values of the cross correlation matrix that, when summed has the greatest value, is the preferred reference channel. The correlation values to be used are those between this reference channel a d the other channels . If we think of the signal and noise as vectors, this diagram illustrates their relationship when the noise is orthogonal to the signal.

Noise

Signal

From the diagram, we can write the SNR as :

Signal 1 cos (θ)

SNR =

Noise tan©) sin(θ)

We assume the peak correlation between two channels goes as p= cos (θ) . This means that when two channels are perfectly correlated, p= 1. This happens when there is no noise, or θ = 0 and the SNR is infinite. Then we can write the SNR in terms of p, the peak correlation between channels :

SNR P

We cannot use this estimate for the reference channel because its correlation with itself is p= 1, which gives infinite SNR. Instead, we use the peak correlation value between the reference and the channel with the highest correlation with the reference. The weight for the reference channel is then set to co = 1. The weights of each other channel are set proportionally, according to the relationship derived previously of its SNR to the reference channel SNR. Finally, the acoustic signals from the n channels are combined using the weighted sum:

Σ cc_ix₁ i=l

where: y β optimally weighted sum of the channels, i = measurement from ith channel, α>i = weight for ith channel .

This method allows an estimate of the beamformer output without reference to a particular location in space. In addition, by using the lags found before an intervention with the sensors at the same physical locations after an intervention, the beam can look at the same location as before the intervention without knowing speci ically where it is . There is no requirement for knowledge of the velocity in the media. A similar method can be developed for the frequency domain. The conventional beamformer output is a Rayleigh quotient . The steering vector that will produce the greatest output is the first eigenvector of the R matrix and its value is the first eigenvalue of the R matrix: Then the output of the beam former is:

S = eVj_. is the first eigenvector of the R matrix. Then the output of the beam former is : e Re V. output in power __ tf __ ^υ 1 ev_x ev_x

So the maximum output of the beamformer can be determined with out knowledge of the velocities in the media. The eigenvector can be saved and after intervention if the sensors are in the same place. It can be used to steer the beam to where the maximum output was previously. Since there are as many R matrices as frequencies of interest a plot of the first or dominate eigen values vs . frequency gives an eigen spectrum, which is the magnitude of the sum of the correlated parts of the channels at each frequency.

Claims

I CLAIM:

1. An acoustic sensor array for the detection of blood flow sounds wherein a plurality of sensors define an array aperture and array geometry, and wherein said array aperture and array geometry are configured such that said plurality of sensors are located within a standardized acoustic window size and wherein said acoustic sensor array is positionable on the chest of a patient.

2. A sensor array according to Claim 1 , wherein said plurality of sensors are positioned on at least four linear and spaced apart PVDF strips which are configured to transversely extend across a portion ofthe chest ofthe patient.

3. A sensor array according to Claim 1, wherein said acoustic sensors in said array are piezoelectric film sensors.

4. A sensor array according to Claim 2, wherein said acoustic sensors in said array are piezoelectric film sensors.

5. A sensor array according to Claim 2, wherein said acoustic sensor array comprises at least one linear multi-channel acoustic sensor array.

6. A sensor array according to Claim 1, wherein said acoustic sensor array comprises at least one linear multi-channel acoustic sensor array.

7. A claim 1 array wherein said array comprises a plurality of spaced apart generally linear sensor strip arrays and wherein at least one of said spaced apart generally linear sensor strip arrays can be located above and aligned with a region corresponding to a selected intercostal space when said array is positioned on the chest of a patient.

8. A claim 7 array wherein said sensors in said array comprise at least one of piezoelectric film sensors and polyvinylidene difluoride film sensors.

9. An acoustic sensor array according to Claim 1 , wherein said standardized acoustic window size is based on a statistical correlation of parameters measured over a number of human subjects.

10. A set of acoustic sensor arrays, comprising: a first acoustic sensor array having a first size and a first geometry corresponding to a first population segment; and a second acoustic sensor array having a second size and a second geometry corresponding to a second population segment.

11. A set of acoustic sensor arrays according to Claim 10, wherein said first acoustic sensor array geometry is substantially the same as the second acoustic sensor array geometry.

12. A set of acoustic sensor arrays according to Claim 11, wherein said first acoustic sensor array size is scaled to be smaller than said second acoustic sensor size to thereby more closely fit populations having smaller acoustic windows.

13. A method for determining the acoustic window which comprises the steps of:

(i) determining the perimeters of an acoustic window of an individual; and

(ii) providing an acoustic sensor array having an aperture sized to accommodate said acoustic window perimeter.

14. The claim 13 method further comprising the step of: (iii) positioning within said aperture of said array a plurality of sensors, the number of sensors selected corresponding to sensor size and by the quality ofthe combined signal from sensors associated with an acoustic evaluation.

15. A method for identifying acoustic sensor array configurations or positions, comprising the steps of:

(i) determining the perimeter of a proximate acoustic window area separately for a plurality of adjacent intercostal spaces; and

(ii) merging two or more of said proximate intercostal space window areas, wherein a merged acoustic window region is defined.

16. The claim 15 method further comprising the step of:

(iii) providing a sensor array wherein said array comprises an aperture sized to accommodate said merged acoustic window of step (ii).

17. The claim 16 method wherein said merged window areas are the fourth and fifth intercostal space window areas.

18. A method of identifying acoustic windows suitable for CAD acoustic evaluations, which comprises the steps of:

(i) determining the average size ofthe acoustic window of a plurality of patients; and

(ii) providing at least one acoustic array geometry which accommodates a predetermined number of sensors within said average acoustic window size as determined in step (i).

19. A template for determining at least one suitable or appropriate location with which to align an acoustic sensor array on the chest of a patient, wherein said template includes a perimeter corresponding to the average size ofthe acoustic window of a plurality of individuals, and wherein said template includes indicia to indicate an appropriate position of said template to overlie a person's chest so that an acoustic sensor array can be aligned therewith.

20. A method for identifying the acoustic window of a person which comprises the steps of:

(i) causing said person to assume a supine position;

(ii) drawing a series of indicia along the left sternal border at the beginning of each intercostal space for each of intercostal spaces 1 to 6 of said person;

(iii) placing an ultrasound probe at the left sternal border of each of said intercostal spaces and moving said probe along each said intercostal space until lung tissue is encountered;

(iv) placing an indicator on the chest of said person to identify as to each intercostal space where lung tissue is encountered;

(v) drawing a series of lines connecting said lung indicator indicia to said left sternal border wherein said lines indicate the bounds ofthe acoustic window perimeter of one corresponding intercostal space; and wherein a perimeter enclosing all of said lines defines said perimeters of said total acoustic window area of said person.

21. A method for determining a location associated with an acoustic window which comprises the step of:

(i) separately determining acoustic window regions proximate to a plurality of adjacent intercostal spaces, wherein each of said proximate acoustic window regions comprises a region of an intercostal space extending from the left sternal border to a point above lung tissue.

22. A method for determining an acoustic window suitable for passive- acoustic coronary artery disease evaluation, comprising the steps of: (a) positioning a multi-channel acoustic sensor array onto the chest of a subject;

(b) calculating a weighted signal value for each ofthe sensor channels in the multi-channel sensor array;

(c) determining the location of each sensor corresponding to the sensor channel in the array;

(d) identifying the sensor channels which meet predetermined test criteria; and

(e) defining a perimeter which substantially extends about and encloses therewithin the sensors associated with the sensor channels identified in step (d), thereby defining an acoustic window suitable for acoustic listening diagnostic procedures.

23. A method according to Claim 22, wherein said multi-channel sensor array comprises at least four separate sensors and a corresponding number of sensor channels.

24. A method according to Claim 22, wherein said calculating step is performed by assigning SNR based weighted values to each ofthe sensor channels.

25. A method according to Claim 22, wherein the predetermined test criteria includes absolute criteria.

26. A method according to Claim 22, wherein the absolute criteria comprises predetermined minimum threshold values.

27. A method according to Claim 22, wherein the predetermined test criteria includes relative criteria.

28. A method according to Claim 27, wherein the relative criteria includes identifying the sensors corresponding to the sensor channels having the three highest weight values.

29. A method according to Claim 23, wherein the predetermined test criteria is defined such that it identifies at least three sensors.

30. A method according to Claim 22, wherein the predetermined test criteria comprises identifying channels exhibiting weight values below a minimum value.

31. A method according to Claim 22, wherein the acoustic window is drawn to include a subset ofthe sensors in the multi-channel sensor array.

32. A method according to Claim 22, wherein the method is repeated across a population of subjects, and wherein the location ofthe sensors associated with the sensor channels identified in step (d) are correlated across the population to define a standardized acoustic window representation.

33. A method according to Claim 22, further comprising the step of configuring at least one standardized multi-channel acoustic sensor array having at least four sensors and a known geometric configuration such that its sensors are positioned substantially within the acoustic window defined by the method of Claim 19.

34. A method according to Claim 33, wherein the multi-channel acoustic sensor array comprises greater than four sensors, and wherein said array has a perimeter and an aperture and wherein said array perimeter substantially conforms to the perimeter identified in the method of Claim 1.

35. A method according to Claim 34, wherein the array and acoustic window perimeter is configured such that it extends about the left ofthe third intercostal space, follows the sternum covering the fourth through six intercostal spaces, and widens to the right at the sixth intercostal space of a person.

36. A method for identifying an acoustic window of a person comprising the steps of:

(a) positioning a multi-channel acoustic sensor array on the chest of a subject;

(b) mapping the optimal weights scaled to an estimated SNR of each ofthe channels ofthe multiple channel acoustic sensor array to the locations of each sensor in said array, wherein said array comprises at least four sensors;

(c) identifying a plurality ofthe sensors that correspond to the sensor channels that have high optimal weights, the plurality of sensors being less than all ofthe sensors in said array; and

(d) identifying a perimeter enclosing the sensors associated with the sensor channels identified in step (c) to define the bounds of an acoustic window for acoustic diagnostic evaluation ofthe subject, wherein the sensors associated with the sensor channels identified in step (c) are associated with a plurality of chest locations ofthe subject.

37. A method according to Claim 35, further comprising the step of configuring a multi-channel acoustic sensor array such that the sensors substantially fit within the acoustic window defined by steps (b)-(d).

38. A method according to Claim 37, wherein the multi-channel acoustic sensor array comprises an array aperture, wherein the size and geometry ofthe aperture is determined based on the identified perimeter ofthe acoustic window defined by the method of Claim 35.

39. A method according to Claim 38, wherein the multi-channel acoustic sensor array comprises a plurality of sensors, wherein said array has a perimeter and an aperture and wherein said array perimeter substantially conforms to the bounds of an acoustic window that starts at the left ofthe third intercostal space, follows the sternum covering the fourth through six intercostal spaces, and widens to the right at the sixth intercostal space of a person.