IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL 38, NO. 5 , MAY I991
483
Reconstruction of Sequential Cardiac In-Plane Displacement Patterns on the Chest Wall by Laser Speckle Interferometry Megha Singh and G . Ramachandran
Abstract-Time-average speckle interferometry has been applied to obtain displacement patterns on the chest wall produced by cardiac action, in the absence of breathing, during various phases of the cardiac cycle. This has been achieved by an electronic shutter, controlled by the electrocardiogram of the subject. The recorded holographic plates processed under identical conditions are scanned by the pointwise method to obtain the absolute displacements at various locations corresponding to the activities of the various cardiac chambers and values. These data are transformed to a 40 x 30 matrix by an interpolation method and, from this, three-dimensional displacement plots are reconstructed by an IBM PC/AT computer. These patterns shown the displacements over the entire cardiac area corresponding to the activities of various regions during the cardiac cycle. The apex and aortic valve areas show the maximum displacements during the systolic phase. During the diastolic phase the activities over the low pressure atrial regions are also observed. The results obtained outline the functional details of the normal heart and the activities over various areas are in agreement with that of other noncontact techniques.
INTRODUCTION EASUREMENT of chest wall vibrations over the cardiac region has been of particular interest to scientists for the past several decades. This has resulted in the development of various contact and noncontact techniques [ 11-16]. These techniques have provided useful data on cardiac functioning, but the application of contact type has always resulted in the attenuation of the displacement signals [ 11-[3], whereas noncontact techniques have generally been confined to the measurement over a smaller region of the heart 141-161. Thus, by these methods precise details of the mechanical activity of the entire heart could not be obtained. The appearance of speckles is an interference phenomenon which is always present when an optically rough surface is illuminated by a laser beam 171. With the movement of the surface, the speckle pattern is also displaced. If two speckle patterns produced by initial and final positions of the object are superposed on the holographic
M
Manuscript received July 12, 1989; revised July 23, 1990. This work was supported by the Department of Science and Technology, Government of India, under Grant SP/S2/L22/85. The authors are with the Biomedical Engineering Division, Indian Institute of Technology, Madras 600 36, India. IEEE Log Number 9144700.
plate, the corresponding displacement could be measured by analyzing the resultant pattern by an unexpanded laser beam. This method has successfully been applied for the measurement of in-plane and out-of-plane displacements (ranging from a few microns to a hundred microns), strain and deformation of various surfaces [8], and deformation in the interior of the three-dimensional bodies [9]. Recording these patterns could either be performed by the double-exposure (two exposures-one at rest and the other at the displaced position) or by the time-average (single exposure of a fixed duration covering any two positions over a cycle) method. Holographic and speckle pattem interferometry are versatile optical techniques which can cover the entire cardiac region. In the former method, variation in the displacements could be observed from the displacement contour fringes obtained at various locations, whereas in speckle interferometry, the recorded speckle pattems have to be analyzed by an appropriate method in order to obtain the displacement information [8]. Hok et al. [IO], applying the holographic technique on the chest, have demonstrated definite changes in the fringe pattems obtained from cardiac patients compared to normal subjects, but this method requires a high power ruby laser and is very sensitive to environmental disturbances. Time-average speckle interferometry is a relatively simple technique and the application of this method during the systolic phase of the ECG has shown its capability to measure the displacements on the entire cardiac area [ 1 I]. By suitably timing the exposure of the chest to the laser light using an electronic shutter, changes in the cardiac events during the I and I1 sounds of the PCG 1121, and P-, QRS-, and Twaves of the ECG [ 131 have been obtained and presented in the form of three-dimensional displacement plots. These plots provide details of the variations in the cardiac activity as observed on the chest wall. As such, these patterns may contain information on a combination of events of the cardiac cycle as the exposure times are fairly large. To well define the displacements pertaining to specific events of the heart it is important to split these large durations into small intervals. To achieve this, the entire cardiac cycle could be divided into various intervals using R-wave peak as a reference. To the authors’ knowledge, such details of the displacements are
00 18-9294191/0500-0483$0 I .OO
0 1991 IEEE
484
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING. VOL. 38. NO. 5 . MAY 1991
not available as yet. Therefore, the aim of the present work is to analyze the displacement patterns of the heart, as observed by the movement of the speckles on the chest wall during the sequential intervals of the ECG. As it is difficult to utilize the double exposure method for functional analysis of the heart, time-average speckle interferometry (TASI) with a He-Ne laser of power 20 mW has been utilized.
METHODS Illumination of the chest wall of the subject by a collimated beam of laser, recording of the speckle patterns during the required phases of the ECG by using an electronic shutter, and analysis of the recorded plates to obtain the displacements and generation of 3-D plots from these data are the various steps involved in the present analysis [ 111. Illumination A beam of He-Ne laser (A = 632.8 nm, power = 20 mW) was spatially filtered and collimated. Fig. I shows the area above the cardiac region illuminated by this collimated beam which was initially determined by an X-ray in the PA-position. This area consists of the chest wall from I1 to V intercostal space with the sternum at the right side. The activity at various locations, which was maximum at the apex region, could easily be observed by the movement of speckles in the absence of breathing. To improve the reflectivity of this area a thin layer of water paint was applied. Shutter Operation The ECG of the subject was continuously monitored and was used to operate a shutter placed in the path of the laser beam to allow illumination only during the required phase. The circuit diagram of the shutter is given in Fig. 2(a). The ECG was fed to a filter and comparator to sense the R-wave peak. The output from the comparator in synchronism with the R-wave peak was fed to a pulse programmer (Yamuna Digital Electronics, India, Model 104DO). The output pulse was supplied to a relay to allow the passage of the laser beam for the required duration during the desired phase of ECG by varying its delay time and width. A further check on this operation was performed by placing a light-dependent resistor after the shutter [Fig. 2(b)]. This output was also displayed along with the ECG of the subject. Recording The complete recording assembly which was placed on a vibration-free table is shown in Fig. 3 . The subject was asked to sit in a vibration-free chair and was advised to hold his breathing during recording, which was further checked by two air transducer operating at frequency 36 kHz [ 111. This was to ensure that the chest and abdominal wall did not suffer any respiratory movement. An in-plane
Fig. I . Location on the chest wall for recording the specklegram
4*-o-
a
483
Reconstruction of Sequential Cardiac In-Plane Displacement Patterns on the Chest Wall by Laser Speckle Interferometry Megha Singh and G . Ramachandran
Abstract-Time-average speckle interferometry has been applied to obtain displacement patterns on the chest wall produced by cardiac action, in the absence of breathing, during various phases of the cardiac cycle. This has been achieved by an electronic shutter, controlled by the electrocardiogram of the subject. The recorded holographic plates processed under identical conditions are scanned by the pointwise method to obtain the absolute displacements at various locations corresponding to the activities of the various cardiac chambers and values. These data are transformed to a 40 x 30 matrix by an interpolation method and, from this, three-dimensional displacement plots are reconstructed by an IBM PC/AT computer. These patterns shown the displacements over the entire cardiac area corresponding to the activities of various regions during the cardiac cycle. The apex and aortic valve areas show the maximum displacements during the systolic phase. During the diastolic phase the activities over the low pressure atrial regions are also observed. The results obtained outline the functional details of the normal heart and the activities over various areas are in agreement with that of other noncontact techniques.
INTRODUCTION EASUREMENT of chest wall vibrations over the cardiac region has been of particular interest to scientists for the past several decades. This has resulted in the development of various contact and noncontact techniques [ 11-16]. These techniques have provided useful data on cardiac functioning, but the application of contact type has always resulted in the attenuation of the displacement signals [ 11-[3], whereas noncontact techniques have generally been confined to the measurement over a smaller region of the heart 141-161. Thus, by these methods precise details of the mechanical activity of the entire heart could not be obtained. The appearance of speckles is an interference phenomenon which is always present when an optically rough surface is illuminated by a laser beam 171. With the movement of the surface, the speckle pattern is also displaced. If two speckle patterns produced by initial and final positions of the object are superposed on the holographic
M
Manuscript received July 12, 1989; revised July 23, 1990. This work was supported by the Department of Science and Technology, Government of India, under Grant SP/S2/L22/85. The authors are with the Biomedical Engineering Division, Indian Institute of Technology, Madras 600 36, India. IEEE Log Number 9144700.
plate, the corresponding displacement could be measured by analyzing the resultant pattern by an unexpanded laser beam. This method has successfully been applied for the measurement of in-plane and out-of-plane displacements (ranging from a few microns to a hundred microns), strain and deformation of various surfaces [8], and deformation in the interior of the three-dimensional bodies [9]. Recording these patterns could either be performed by the double-exposure (two exposures-one at rest and the other at the displaced position) or by the time-average (single exposure of a fixed duration covering any two positions over a cycle) method. Holographic and speckle pattem interferometry are versatile optical techniques which can cover the entire cardiac region. In the former method, variation in the displacements could be observed from the displacement contour fringes obtained at various locations, whereas in speckle interferometry, the recorded speckle pattems have to be analyzed by an appropriate method in order to obtain the displacement information [8]. Hok et al. [IO], applying the holographic technique on the chest, have demonstrated definite changes in the fringe pattems obtained from cardiac patients compared to normal subjects, but this method requires a high power ruby laser and is very sensitive to environmental disturbances. Time-average speckle interferometry is a relatively simple technique and the application of this method during the systolic phase of the ECG has shown its capability to measure the displacements on the entire cardiac area [ 1 I]. By suitably timing the exposure of the chest to the laser light using an electronic shutter, changes in the cardiac events during the I and I1 sounds of the PCG 1121, and P-, QRS-, and Twaves of the ECG [ 131 have been obtained and presented in the form of three-dimensional displacement plots. These plots provide details of the variations in the cardiac activity as observed on the chest wall. As such, these patterns may contain information on a combination of events of the cardiac cycle as the exposure times are fairly large. To well define the displacements pertaining to specific events of the heart it is important to split these large durations into small intervals. To achieve this, the entire cardiac cycle could be divided into various intervals using R-wave peak as a reference. To the authors’ knowledge, such details of the displacements are
00 18-9294191/0500-0483$0 I .OO
0 1991 IEEE
484
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING. VOL. 38. NO. 5 . MAY 1991
not available as yet. Therefore, the aim of the present work is to analyze the displacement patterns of the heart, as observed by the movement of the speckles on the chest wall during the sequential intervals of the ECG. As it is difficult to utilize the double exposure method for functional analysis of the heart, time-average speckle interferometry (TASI) with a He-Ne laser of power 20 mW has been utilized.
METHODS Illumination of the chest wall of the subject by a collimated beam of laser, recording of the speckle patterns during the required phases of the ECG by using an electronic shutter, and analysis of the recorded plates to obtain the displacements and generation of 3-D plots from these data are the various steps involved in the present analysis [ 111. Illumination A beam of He-Ne laser (A = 632.8 nm, power = 20 mW) was spatially filtered and collimated. Fig. I shows the area above the cardiac region illuminated by this collimated beam which was initially determined by an X-ray in the PA-position. This area consists of the chest wall from I1 to V intercostal space with the sternum at the right side. The activity at various locations, which was maximum at the apex region, could easily be observed by the movement of speckles in the absence of breathing. To improve the reflectivity of this area a thin layer of water paint was applied. Shutter Operation The ECG of the subject was continuously monitored and was used to operate a shutter placed in the path of the laser beam to allow illumination only during the required phase. The circuit diagram of the shutter is given in Fig. 2(a). The ECG was fed to a filter and comparator to sense the R-wave peak. The output from the comparator in synchronism with the R-wave peak was fed to a pulse programmer (Yamuna Digital Electronics, India, Model 104DO). The output pulse was supplied to a relay to allow the passage of the laser beam for the required duration during the desired phase of ECG by varying its delay time and width. A further check on this operation was performed by placing a light-dependent resistor after the shutter [Fig. 2(b)]. This output was also displayed along with the ECG of the subject. Recording The complete recording assembly which was placed on a vibration-free table is shown in Fig. 3 . The subject was asked to sit in a vibration-free chair and was advised to hold his breathing during recording, which was further checked by two air transducer operating at frequency 36 kHz [ 111. This was to ensure that the chest and abdominal wall did not suffer any respiratory movement. An in-plane
Fig. I . Location on the chest wall for recording the specklegram
4*-o-
a
