Ischemic heart disease is ranked 1 for the total cause of deaths in the world. 9.14 million people died due to ischemic heart disease in 2019, which makes up 16.7% of all deaths in 2019. Myocardial Infarction, also known as heart attack, is one of the most significant type of ischemic heart disease causing deaths. STEMI is the most severe type of Myocardial Infarction that causes death or disability. A typical STEMI diagnosis is done with 12-lead ECG and it has to be quick, in 10 minutes from the first medical contact, in order to save the patient. However, previous studies among physicians and paramedics have shown that the accuracy of STEMI diagnosis by the 12-lead ECG is not sufficient. Thus, this thesis is focused on implementing an artificial intelligence model that can detect and locate if the 12-lead ECG image has STEMI signals or not. The dataset for this thesis consists of STEMI and not-STEMI 12-lead ECG images. The dataset is provided by Hualien Tzu Chi Hospital. 2D-CNN model is proposed for this thesis. The model has 10 layers with less than 13000 parameters in total which makes the inference time fast. The 2D-CNN model is trained with 540 ECG images and it achieved 96.3% accuracy, 96.2% sensitivity, 89.4% precision, 0.926 F1-score, and 0.962 ROC-AUC score for 537 testing images. The thesis model is compared to 10 different transfer learning models. The thesis model has the best accuracy, sensitivity, F1-score and ROC-AUC score. Grad-CAM technique is used for localization of STEMI signals in ECG images. According to the comparisons, the thesis model is the most reliable for localizing the STEMI signals. This localization builds trust for the model because the CNN model is not a black box anymore, we can see where the CNN model looks and decides. The result of localization can also be used for teaching inexperienced physicians and paramedics. As a result, the proposed model is effective to detect and locate STEMI signal in a 12-lead ECG image. Also, the proposed model would be helpful for accurate diagnosis of STEMI in a short time for clinical practices.

1 Introduction 1
1.1 Motivation 5
1.2 Organization of Thesis 5
2 Background 7
2.1 Artificial Intelligence 7
2.1.1 Classification 8
2.2 Neural Networks 8
2.2.1 Multi-Layer Perceptron 9
2.2.2 Activation Functions 10
2.2.3 Training Process 11
2.2.4 Convolutional Neural Network 12
2.3 Performance Evaluation Metrics 14
3 Related work 17
4 Material and Methods 19
4.1 Dataset 19
4.2 Public Datasets 20
4.3 Pre-processing 20
4.4 The Proposed Architecture 25
5 Results 27
5.1 Using the Neural Network 27
5.2 Using Transfer Learning 28
5.3 Visual Explanation of Models 30
6 Conclusion 33
References 35

[1] G. A. Roth, G. A. Mensah, C. O. Johnson, G. Addolorato, E. Ammirati, L. M. Baddour, N. C. Barengo, A. Z. Beaton, E. J. Benjamin, C. P. Benziger, et al., “Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the gbd 2019 study,” Journal of the American College of Cardiology, vol. 76, no. 25, pp. 2982–3021, 2020.
[2] B. Ibanez, S. James, S. Agewall, M. J. Antunes, C. Bucciarelli-Ducci, H. Bueno, A. L. Caforio, F. Crea, J. A. Goudevenos, S. Halvorsen, et al., “2017 esc guidelines for the management of acute myocardial infarction in patients presenting with st-segment elevation: The task force for the management of acute myocardial infarction in patients presenting with st-segment elevation of the european society of cardiology (esc),” European heart journal, vol. 39, no. 2, pp. 119–177, 2018.
[3] A. Hoang, A. Singh, and A. Singh, “Comparing physicians and experienced advanced practice practitioners on the interpretation of electrocardiograms in the emergency department,” The American Journal of Emergency Medicine, vol. 40, pp. 145–147, 2021.
[4] J. M. McCabe, E. J. Armstrong, I. Ku, A. Kulkarni, K. S. Hoffmayer, P. D. Bhave, S. W. Waldo, P. Hsue, J. C. Stein, G. M. Marcus, et al., “Physician accuracy in interpreting potential st-segment elevation myocardial infarction electrocardiograms,” Journal of the American Heart Association, vol. 2, no. 5, p. e000268, 2013.
[5] F. Rosenblatt, “The perceptron: a probabilistic model for information storage and organization in the brain.,” Psychological review, vol. 65, no. 6, p. 386, 1958.
[6] G. Cybenko, “Approximation by superpositions of a sigmoidal function,” Mathematics of control, signals and systems, vol. 2, no. 4, pp. 303–314, 1989.
[7] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by backpropagating errors,” nature, vol. 323, no. 6088, pp. 533–536, 1986.
[8] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998.
[9] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Advances in neural information processing systems, vol. 25, pp. 1097–1105, 2012.
[10] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014.
[11] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770– 778, 2016.
[12] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Imagenet: A large-scale hierarchical image database,” in 2009 IEEE conference on computer vision and pattern recognition, pp. 248–255, Ieee, 2009.
[13] M. Sharma, R. San Tan, and U. R. Acharya, “A novel automated diagnostic system for classification of myocardial infarction ecg signals using an optimal biorthogonal filter bank,” Computers in biology and medicine, vol. 102, pp. 341–356, 2018.
[14] U. R. Acharya, H. Fujita, V. K. Sudarshan, S. L. Oh, M. Adam, J. E. Koh, J. H. Tan, D. N. Ghista, R. J. Martis, C. K. Chua, et al., “Automated detection and localization of myocardial infarction using electrocardiogram: a comparative study of different leads,” Knowledge-Based Systems, vol. 99, pp. 146–156, 2016.
[15] L. D. Sharma and R. K. Sunkaria, “Inferior myocardial infarction detection using stationary wavelet transform and machine learning approach,” Signal, Image and Video Processing, vol. 12, no. 2, pp. 199–206, 2018.
[16] J. T.-Y. Weng, J.-J. Lin, Y.-C. Chen, and P.-C. Chang, “Myocardial infarction classification by morphological feature extraction from big 12-lead ecg data,” in Pacific-Asia Conference on Knowledge Discovery and Data Mining, pp. 689–699, Springer, 2014.
[17] A. K. Dohare, V. Kumar, and R. Kumar, “Detection of myocardial infarction in 12 lead ecg using support vector machine,” Applied Soft Computing, vol. 64, pp. 138–147, 2018.
[18] R. Tao, S. Zhang, X. Huang, M. Tao, J. Ma, S. Ma, C. Zhang, T. Zhang, F. Tang, J. Lu, et al., “Magnetocardiography-based ischemic heart disease detection and localization using machine learning methods,” IEEE Transactions on Biomedical Engineering, vol. 66, no. 6, pp. 1658–1667, 2018.
[19] U. B. Baloglu, M. Talo, O. Yildirim, R. San Tan, and U. R. Acharya, “Classification of myocardial infarction with multi-lead ecg signals and deep cnn,” Pattern Recognition Letters, vol. 122, pp. 23–30, 2019.
[20] T. Reasat and C. Shahnaz, “Detection of inferior myocardial infarction using shallow convolutional neural networks,” in 2017 IEEE Region 10 Humanitarian Technology Conference (R10-HTC), pp. 718–721, IEEE, 2017.
[21] W. Liu, M. Zhang, Y. Zhang, Y. Liao, Q. Huang, S. Chang, H. Wang, and J. He, “Realtime multilead convolutional neural network for myocardial infarction detection,” IEEE journal of biomedical and health informatics, vol. 22, no. 5, pp. 1434–1444, 2017.
[22] K. Feng, X. Pi, H. Liu, and K. Sun, “Myocardial infarction classification based on convolutional neural network and recurrent neural network,” Applied Sciences, vol. 9, no. 9, p. 1879, 2019.
[23] Y. Cao, T. Wei, N. Lin, D. Zhang, and J. J. Rodrigues, “Multi-channel lightweight convolutional neural network for remote myocardial infarction monitoring,” in 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), pp. 1–6, IEEE, 2020.
[24] L. Fu, B. Lu, B. Nie, Z. Peng, H. Liu, and X. Pi, “Hybrid network with attention mechanism for detection and location of myocardial infarction based on 12-lead electrocardiogram signals,” Sensors, vol. 20, no. 4, p. 1020, 2020.
[25] C. Han and L. Shi, “Ml–resnet: A novel network to detect and locate myocardial infarction using 12 leads ecg,” Computer methods and programs in biomedicine, vol. 185, p. 105138, 2020.
[26] U. R. Acharya, H. Fujita, S. L. Oh, Y. Hagiwara, J. H. Tan, and M. Adam, “Application of deep convolutional neural network for automated detection of myocardial infarction using ecg signals,” Information Sciences, vol. 415, pp. 190–198, 2017.
[27] H. Makimoto, M. Höckmann, T. Lin, D. Glöckner, S. Gerguri, L. Clasen, J. Schmidt, ¨ A. Assadi-Schmidt, A. Bejinariu, P. Müller, et al., “Performance of a convolutional neural network derived from an ecg database in recognizing myocardial infarction,” Scientific reports, vol. 10, no. 1, pp. 1–9, 2020.
[28] S. Hong, Y. Zhou, J. Shang, C. Xiao, and J. Sun, “Opportunities and challenges of deep learning methods for electrocardiogram data: A systematic review,” Computers in Biology and Medicine, p. 103801, 2020.
[29] Y. Park, I. D. Yun, and S.-H. Kang, “Preprocessing method for performance enhancement in cnn-based stemi detection from 12-lead ecg,” IEEE Access, vol. 7, pp. 99964–99977, 2019.
[30] A. L. Goldberger, L. A. Amaral, L. Glass, J. M. Hausdorff, P. C. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,” circulation, vol. 101, no. 23, pp. e215–e220, 2000.
[31] F. Jager, A. Taddei, G. B. Moody, M. Emdin, G. Antolič, R. Dorn, A. Smrdel, C. Marchesi, and R. G. Mark, “Long-term st database: a reference for the development and evaluation of automated ischaemia detectors and for the study of the dynamics of myocardial ischaemia,” Medical and Biological Engineering and Computing, vol. 41, no. 2, pp. 172– 182, 2003.
[32] R. Bousseljot, D. Kreiseler, and A. Schnabel, “Nutzung der ekg-signaldatenbank cardiodat der ptb über das internet,” 1995.
[33] L. Torrey and J. Shavlik, “Transfer learning,” in Handbook of research on machine learning applications and trends: algorithms, methods, and techniques, pp. 242–264, IGI global, 2010.
[34] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” CoRR, vol. abs/1610.02357, 2016.
[35] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the inception architecture for computer vision,” 2015.
[36] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. Alemi, “Inception-v4, inception-resnet and the impact of residual connections on learning,” 2016.
[37] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, “Mobilenetv2: Inverted residuals and linear bottlenecks,” 2019.
[38] G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” 2018.
[39] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in Proceedings of the IEEE international conference on computer vision, pp. 618–626, 2017.