Abstract: Coronary artery disease occurs when plaque is accumulated in the walls of the artery. This causes the artery to narrow, reducing blood flow to the heart. Coronary artery disease is globally identified as the most predominant and lethal cardiovascular disease. Furthermore, undiagnosed coronary artery disease may progress and lead to complications such as myocardial infarction and congestive heart failure. Hence there is a compelling need for the prompt and unerring detection of coronary artery disease, myocardial infarction, and congestive heart failure using automated systems. The electrocardiogram (ECG) is the most preferred method of detecting cardiovascular diseases as it is easily available and economical compared to imaging methods. Hence, this talk summarises the development of advanced models using ECG signals for the detection of coronary artery disease, myocardial infarction, and congestive heart failure, focusing on the detection of myocardial infarction. This work contributes to the medical field as it offers some level of explainability of the inner workings of the deep models that clinicians may relate to. The reliability of the developed deep model used in healthcare applications such as emergency diagnosis of different types of myocardial infarction contributes significantly to clinicians. The presentation will include 3 parts:
1) the development of convolutional neural network (CNN) and GaborCNN (with a unique Gabor layer) models for rapidly classifying coronary artery disease, myocardial infarction, congestive heart failure, and healthy ECG signals is discussed. The ECG signals which were acquired from the Physikalisch- Technische Bundesanstalt (PTB) database, were fed to the two models for classification. The GaborCNN was affirmed to be the better model for the classification task due to its high overall accuracy of 98.74% and lower computational demand. We believe this is the first study to integrate the Gabor filter into the CNN model to automatically classify normal, coronary artery disease, myocardial infarction, and congestive heart failure classes using ECG signals.
Despite the surge in the development of robust models for the automated detection of cardiovascular diseases, these are often not trusted by clinicians due to the lack of explainability of models’ mechanisms. Hence,
2) the development of the CNN and DenseNet models with the application of an advanced and unique GRAD-CAM technique to both models’ output will be briefly discussed. ECG beats were extracted from the healthy and ten myocardial infarction classes using the R peak detection algorithm and fed to the developed CNN and DenseNet models. Application of the GRAD-CAM technique enabled visualization of ECG leads and portions of ECG waves that influenced the models’ predictive decisions. DenseNet was identified as a better model due to its low computational complexity and higher classification accuracy of 98.9% due to feature reusability. Lead V4 was the most activated lead in both models. The DenseNet model with the Grad-CAM technique enables clinicians to determine the type of myocardial infarction based on explainability and, thus, has the potential to boost clinicians’ confidence in using it in hospital settings. This is exciting to report features that influenced the classification decisions of deep models for multiclass classification of myocardial infarction and healthy ECGs.
Current diagnostic models for cardiovascular diseases have been primarily developed using public databases and are thus unsuitable for hospital settings, where the uncertainty of models is predominant.
3) a unique Dirichlet DenseNet model was trained with pre-processed myocardial infarction ECG signals and tested with noisy myocardial infarction signals. The predictive entropy was used as an uncertainty measure to determine the misclassification of normal and myocardial infarction signals. The misclassification of signals was determined based on the computation of four uncertainty metrics; uncertainty sensitivity, specificity, accuracy and precision. The proposed method demonstrates that the developed model is reliable as it is able to convey when it is not confident in the diagnostic information its presenting, having the potential to make a significant contribution to clinicians, especially in emergencies such as urgent diagnosis of myocardial infarction. We have explored uncertainty quantification of a deep model using multi-class myocardial infarction ECG signals.
In summary, we believe the models proposed in the above 3 parts have great potential to contribute significantly to healthcare in areas such as the emergency diagnosis of acute myocardial infarction.

Biography: Eddie is elected as:
* Academician for European Academy of Sciences (FEUAS, Greece);
* Fellow (inaugural) for National Academy of Technology (FNAT, USA);
*Academician for European Academy of Sciences and Arts (AEASA, EU-Austria);
* Fellow of the American Society of Mechanical Engineers (FASME, USA);
* Fellow of Institute of Engineering and Technology (FIET, United Kingdom);
* Fellow of International Engineering and Technology Institute (FIETI, Hong Kong);
* Distinguished Fellow for Institute of Data Science and Artificial Intelligence (DFIDSAI, China);
* Academician for Academy of Pedagogy and Learning (USA).

He has published numerous papers in SCI-IF int. journal (530); int. conf. proceedings (140), textbook chapters (>110) and others (32) over the 31 years. Co-edited 14 books in STEM areas.

He is in the Stanford list of the World’s top 2% Scientists since 2019 (ranked 173 as top 0.001% in the field of Biomedical Engineering), and ranked # 6 (Worldwide) in Google Scholar under Biomedical, cited by 18,050 (h-index: 68).

He is the:
* Lead Editor-in-Chief for the ISI Journal of Mechanics in Medicine and Biology for dissemination of original research in all fields of mechanics in medicine and biology since 2000;
* Founding Editor-in-Chief for the ISI indexed Journal of Medical Imaging and Health Informatics (2011-2021);
* Associate editor or EAB of various referred international journals such as Applied Intelligence, BioMedical Engineering OnLine, Sensors, Computers in Biology & Medicine, and, Journal of Advanced Thermal Science Research.
More details can be found in: Cv: https://dr.ntu.edu.sg/cris/rp/rp00847.

Ng obtained Ph.D. at Cambridge Univ. and elected as an Academician for European Academy of Sciences and Arts (Austria); Academician for European Academy of Sciences (Greece); a Fellow of The American Society of Mechanical Engineers; The Institution of Engineering and Technology [UK], and International Engineering & Technology Institute [HK]. He researches in numerical simulation in the biomedical engineering, thermal-fluids and health-related diagnosis fields. He is Editor-in-Chief for 2 ISI-journals which were captured by the JCR within 2-years of their inauguration. He has been recognized internationally for academic excellence. He received numerous best papers, service awards and has graduated 26 PhD and 26 Master students. He was awarded the SPRING-Singapore Merit Award for his work in thermal imagers to screen SARS fever and contributions to the Singapore Standardization Program. Twenty-one of his papers have been adopted as references in Singapore Standard (SS-582, Parts 1&2: 2020) and ISO/IEC 80601-2-59: 2017. He serves as a panel member for Singapore Biomedical and Health Standards Committee since 2011. Being a co-inventor of 3 US patents on software classifiers to identify the different stages of breast cancer development in iTBra-system, he was accoladed with equity in a listed company. His ongoing work on non-contact screening for carotid artery stenosis, superficial vein-finder and dual-point Photoplethysmogram (2PPG) has resulted in 4 TDs and another patent on “IoT enabled EPCG-device-unit for nursing heart-health-distantly” with AusPat. Office. He has notable citations in the field of infrared physics & technology.