Dual analysis of myocardial infarction using fractional mathematical modeling and machine learning

Gunasekar, Tharmalingam; Banu, Sumaiya; Santra, Shyam Sundar; Baleanu, Dumitru

doi:10.22124/jmm.2025.31534.2837

Dual analysis of myocardial infarction using fractional mathematical modeling and machine learning

Articles in Press

Document Type : Research Article

Authors

¹ Department of Mathematics, Vel Tech Rangarajan Dr. Sagunthala R\&D Institute of Science and Technology, Chennai - 600062, Tamil Nadu, India

² JIS College of Engineering, Kalyani, Nadia, West Bengal

³ Department of Computer Science and Mathematics, Lebanese American University, Beirut - 11022801, Lebanon

10.22124/jmm.2025.31534.2837

Abstract

This paper presents a novel fractional-order mathematical model of myocardial infarction in women who are users of combined oral contraceptive pill and who also develop comorbidity due to various reasons. The system of equations incorporate Caputo fractional derivative to capture memory effects of the model. Existence and uniqueness of solution of the mathematical model is derived. Numerical simulations were rigorously conducted on the math model with varying fractional order namely, $0.3$, $0.5$ and $0.8$ using Euler's method. The numerical results thus obtained are simulated by Adam's method for 200 days period. The output from these simulations form the dataset of the Bayesian regularization neural network (BRNN) with dataset split for training, testing and validatating the computational model. Bayesian regularization is incorporated to handle overfitting efficiently. Root Mean Square Error (RMSE) are computed for all three fractional orders respectively. Regression analysis is conducted which yielded perfect correlation $(R=1)$ accross the all datasets. The combined mathematical and computational analysis form a strong layout in myocardial infarction risk prediction, diagnosis and treatment in young women.

Keywords