Paper Title
MERGING ARTIFICIAL INTELLIGENCE WITH DIGITAL TWIN MODELS FOR CLASSIFYING CARDIAC ISCHEMIA, FRAMEWORK OF CARDIAC CONDITIONS CLASSIFICATION

Abstract
Abstract - Even if advances in intelligent medical care have been made, many issues still exist. Designing, planning, improving and controlling the cardiac system is still very challenging due to the system’s complexity governed. There is a lack of intelligent supervision, including intelligent monitoring, feedback, and management. Existing systems or platforms do not provide continuous personal health management services throughout a patient’s lifecycle. Patient crisis warning services are inaccurate enough. Routine real-time communication between healthcare providers and patients is not present. There is no true integration of physical medical systems with intelligent information systems. In this paper, a Hybrid Cardiac Twin (HCT) framework is proposed along with an Artificial Neural Network (ANN) to solve issues of real-time supervision with high accuracy in decision-making and prediction of the future crisis. A case study of cardiac ischemia is tested to demonstrate the feasibilities of HCT with a Training, Validation, and Testing regression accuracy of 0.99997, 0.99954, and 0.99931respectively, and the overall accuracy is 0.99986. The proposed framework of HCT and the construction of the ANN model can capture historical and real-time data and manage and predict current and future cardiac conditions. Keywords - Digital Twin, Hybrid Twin, IoT Healthcare, and Machine Learning