Myocardial Infarction Detection using ECG.pptx

Brief Methodology of detecting a myocardial Infarction using ECG.

1. Detection of Myocardial Infarction using single lead ECG Submitted By : Project Supervisor Rudra Narayan Dash(2001106417) Dr. Kanhu Charan Bhuyan Rishikant Sahu (2001106414) Lanin Pradhan(2001106393) Jyoti Nanda(2001106390)

2. Content 1. Introduction 2. Research Significance 3. Data Acquisition and Processing 4. Machine Learning Model 5. Results 6. Conclusion and Future Work

3. Introduction Background In this project, we develop a system that takes the ECG signal of a patient, removes the noise affecting the recording, decompose each recording into sub-bands and then, post extraction of statistical features,we train a machine learning base classifier, with the features as the training data. Finally, using the classifier, we predict whether the patient is suffering from a heart attack or not. Purpose To explore the ability of a single-lead ECG to accurately detect myocardial infarction, enhancing rapid diagnosis and treatment interventions. Methodology The study combines data acquisition techniques with advanced signal processing and machine learning models, specifically using LSTM networks, to analyze ECG signals.

4. Myocardial Infarction Myocardial Infarction represents a critical medical emergency characterized by the abrupt and often catastrophic restriction or occlusion of blood supply to a segment of the cardiac muscle. This dire scenario unfolds as a consequence of the formation and subsequent lodging of a blood clot within one of the coronary arteries, pivotal conduits responsible for delivering oxygen-rich blood to the heart tissue.

5. Electrocardiogram Electrocardiogram or ECG (also known as EKG) is a recording of a heart’s electrical activity through repeated cardiac cycles. It is usually represented as a graph of voltage versus time of the electrical activity of the heart using electrodes placed on the skin. These electrodes detect the small electrical changes that are a consequence of cardiac muscle depolarization followed by repolarization during each cardiac cycle (heartbeat).

6. Standard ECG for a single heart beat Fig 1 : A standard ECG signal for a single heart beat

7. ST-Elevation Myocardial Infarction STEMI, or ST elevation myocardial infarction, stands as a critical manifestation of cardiovascular disease, representing a dire scenario wherein one of the coronary arteries, responsible for supplying oxygen-rich blood to the heart muscle, becomes acutely obstructed. This occlusion often stems from the rupture of an atherosclerotic plaque, precipitating the formation of a blood clot that swiftly impedes blood flow.

8. Non ST-Elevation Myocardial Infarction Non-ST elevation myocardial infarction is a specific type of heart attack characterized by the partial blockage or narrowing of one or more coronary arteries, leading to a decrease in blood flow to a section of the heart muscle. In contrast to ST elevation myocardial infarction, where there is a complete blockage of a coronary artery, NSTEMI involves a milder obstruction, often caused by the accumulation of fatty deposits within the artery walls, known as atherosclerosis.

9. Research Significance Consequences of Delayed Detection Delayed diagnosis of Myocardial Infarction (MI) can lead to severe complications including cardiogenic shock and death. Early detection is crucial for immediate treatment. Benefits of Single Lead ECG Single lead ECGs are more portable and easier to use than traditional multi-lead systems, allowing quicker and more accessible evaluations.

10. Data Acquisition and Processing The Analog Front-End Utilizes an AD620 instrumentation amplifier to capture and amplify ECG signals. Signal Processing Techniques Incorporates Hamilton-Tompkins segmentation and Fourier Decomposition for noise reduction and signal clarification.

11. Machine Learning Model Use of LSTM Networks Long Short-Term Memory (LSTM) networks were trained to classify ECG signals into indicative of MI or not. The system achieved a high accuracy with an AUC score of 0.9402. Statistical Feature Extraction Critical for training the machine learning model by quantifying key aspects of processed ECG segments.

12. Block Diagram of Hamilton Tompkins algorithm Fig 2 : Block Diagram of Hamilton Tompkins algorithm

13. Hardware Model (Schematic)

14. Results Performance Metrics The model showed high precision (87.50%), accuracy (86.61%), recall (86.56%), and F1 score (87.03%). System Efficiency Demonstrated potential for real-time medical diagnostics, enabling timely interventions.

15. Conclusion The study successfully demonstrated that a single-lead ECG system, equipped with advanced signal processing and LSTM-based classification, can effectively detect myocardial infarction.

16. Future Work Future research will focus on improving the model's sensitivity and exploring its adaptability to other cardiovascular conditions.

17. References [1] K. Thygesen, J. S. Alpert, H. D. White, and J. et al., “Universal definition of myocardial infarction,” Circulation, vol. 116, no. 22, pp. 2634–2653, Oct. 2007. [2] D. M. Krikler, “Historical aspects of electrocardiography,” Cardiol Clin, vol. 5, no. 3, pp. 349–355, Aug.1987. [3] L. S. Lilly et al., Pathophysiology of heart disease: a collaborative project of medical students and faculty. Lippincott Williams & Wilkins, 2011. [4] E. A. Ashley and J. Niebauer, Conquering the ECG. London, England: REMEDICA, 2004. [5] K. Thygesen, J. S. Alpert, A. S. Jaffe, M. L. Simoons, B. R. Chaitman, and H. D. White, “Third universal definition of myocardial infarction,” Circulation, vol. 126, no. 16, p. 2020–2035, Oct. 2012. [Online]. Available: http://dx.doi.org/10.1161/CIR.0b013e31826e1058

18. [6] P. G. Steg, E. Bonnefoy, S. Chabaud, F. Lapostolle, P.-Y. Dubien, P. Cristofini, A. Leizorovicz, and P. Touboul, “Impact of time to treatment on mortality after prehospital fibrinolysis or primary angioplasty: Data from the captim randomized clinical trial,” Circulation, vol. 108, no. 23, p. 2851–2856, Dec. 2003. [Online]. Available: http://dx.doi.org/10.1161/01.CIR.0000103122.10021.F2 [7] A. T. Collaboration, “Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients,” Bmj, vol. 324, no. 7329, pp. 71–86, 2002. [8] J. C. B. Ferreira and D. Mochly-Rosen, “Nitroglycerin use in myocardial infarction patients,” Circ J, vol. 76, no. 1, pp. 15–21, Nov. 2011.

19. Thank You