MIT Automated Auscultation System

At every annual exam, the primary care physician uses a stethoscope to listen for cardiac abnormalities. This approach is non-invasive, inexpensive, and fast. It is also highly unreliable. Over 80% of the people referred to cardiologists as suffering from the most commonly diagnosed condition, mitral valve prolapse (MVP), do not have this condition. Working in conjunction with cardiologists at MGH, we developed a robust, low cost, easy to use tool that can be employed to diagnose MVP in the office of primary care physicians. The system fuses signals from an electronic stethoscope and a two-lead EKG, and uses software running on a desktop or laptop computer to make a diagnosis. We also provide a number of novel audiovisual diagnostic aids. These allow physicians to visualize both individual heart beats and a visual-prototypical heart beat constructed from a sequence of beats. They also permit doctors to listen to an audio-prototypical heart-beat, audio enhanced heart-beats that amplify clinically significant sounds, and slowed down heart-beats that make it easier to separate clinically relevant cardiac events. We tested our system on 51 patients. The number of false positives was reduced to approximately 10%. While there is no generally accepted statistic on false negatives, anecdotal experience indicates that our system also outperforms physicians in this respect. Thesis Supervisor: John V. Guttag Title: Professor, Computer Science and Engineering

[1]  D. Thompson,et al.  The Conduction System of the Heart , 1982, Springer Netherlands.

[2]  R Williams,et al.  Computer analysis techniques for phonocardiogram diagnosis. , 1976, Computers and biomedical research, an international journal.

[3]  M Okuni,et al.  Trial of a new cardiac mass screening system in school children: use of portable computer system for heart sound and electrocardiogram--a preliminary report. , 1978, Japanese circulation journal.

[4]  M. Misaki,et al.  Time-scale modification of speech signals using cross-correlation functions , 1992 .

[5]  Jorge Herbert de Lira,et al.  Two-Dimensional Signal and Image Processing , 1989 .

[6]  J L Reynolds,et al.  Heart disease screening of preschool children. Comparison of computer and physical examination methods. , 1970, American journal of diseases of children.

[7]  P. White,et al.  Analysing paediatric heart murmurs with discriminant analysis , 1998, Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vol.20 Biomedical Engineering Towards the Year 2000 and Beyond (Cat. No.98CH36286).

[8]  B. El-Asir,et al.  Time-frequency analysis of heart sounds , 1996, Proceedings of Digital Processing Applications (TENCON '96).

[9]  E. M. Brockbank,et al.  Heart Sounds and Murmurs , 1912, Ugeskrift for laeger.

[10]  J D Frost,et al.  Error reduction in small sample averaging through the use of the median rather than the mean. , 1968, Electroencephalography and clinical neurophysiology.

[11]  S E Kowalski,et al.  Mitral valve prolapse , 1985, Canadian Anaesthetists' Society journal.

[12]  Paul R. White,et al.  Time-frequency analysis of heart murmurs in children , 1997 .

[13]  Francis F. Lee,et al.  Time Compression and Expansion of Speech by the Sampling Method , 1972 .

[14]  G. Fairbanks,et al.  Method for time of frequency compression-expansion of speech , 1954 .

[15]  J. Hertzberg,et al.  Artificial Neural Network-Based Method of Screening Heart Murmurs in Children , 2001, Circulation.

[16]  David Laurenson,et al.  Estimating clean speech thresholds for perceptual based speech enhancement , 1999, Proceedings of the 1999 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. WASPAA'99 (Cat. No.99TH8452).

[17]  Jean Laroche,et al.  New phase-vocoder techniques for pitch-shifting, harmonizing and other exotic effects , 1999, Proceedings of the 1999 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. WASPAA'99 (Cat. No.99TH8452).

[18]  M. Braga,et al.  Exploratory Data Analysis , 2018, Encyclopedia of Social Network Analysis and Mining. 2nd Ed..

[19]  I. Cathers,et al.  Neural network assisted cardiac auscultation , 1995, Artif. Intell. Medicine.

[20]  Mark Dolson,et al.  The Phase Vocoder: A Tutorial , 1986 .

[21]  Bruce Williamson.  Diastole , 1949, Edinburgh Medical Journal.

[22]  Nils J. Nilsson,et al.  Artificial Intelligence , 1974, IFIP Congress.

[23]  E. Craige,et al.  Should auscultation be rehabilitated? , 1988, The New England journal of medicine.

[24]  J Grayzel.  The Cardiac Cycle , 1991, Journal of cardiothoracic and vascular anesthesia.

[25]  Professor Dr. Marc Gertsch.  The ECG , 2004, Springer Berlin Heidelberg.

[26]  W. W. Myint,et al.  An electronic stethoscope with diagnosis capability , 2001, Proceedings of the 33rd Southeastern Symposium on System Theory (Cat. No.01EX460).

[27]  R. N. Rogers,et al.  HEART SOUNDS. , 1964, The Alabama journal of medical sciences.

[28]  Nancy E. Reed,et al.  THE ANALYSIS OF HEART SOUNDS FOR SYMPTOM DETECTION AND MACHINE-AIDED DIAGNOSIS , 2001 .

[29]  A. W. M. van den Enden,et al.  Discrete Time Signal Processing , 1989 .

[30]  L. Nieman,et al.  Cardiac auscultatory skills of internal medicine and family practice trainees. A comparison of diagnostic proficiency. , 1997, JAMA.

[31]  D. Barschdorff,et al.  Automatic phonocardiogram signal analysis in infants based on wavelet transforms and artificial neural networks , 1995, Computers in Cardiology 1995.

[32]  K. Yana,et al.  Detection and classification of systolic murmur for phonocardiogram screening , 1996, Proceedings of 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[33]  L. Nieman,et al.  The Teaching and Practice of Cardiac Auscultation during Internal Medicine and Cardiology Training: A Nationwide Survey , 1993, Annals of Internal Medicine.

[34]  I SUSSMAN,et al.  Heart murmurs. , 2014, Medical times.

[35]  M. Cohen,et al.  Comparative Approaches to Medical Reasoning , 1995 .

[36]  H. T. Nagle,et al.  A comparison of the noise sensitivity of nine QRS detection algorithms , 1990, IEEE Transactions on Biomedical Engineering.

[37]  R. Doraiswami,et al.  Wavelet based bank of correlators approach for phonocardiogram signal classification , 1998, Proceedings of the IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis (Cat. No.98TH8380).