Hidden Markov Models for early detection of cardiovascular diseases

Abstract

Introduction: This article, developed between 2022 and 2023 within the framework of Applied Stochastic Processes by the SciBas group at the Universidad Distrital Francisco José de Caldas, focuses on the role of Hidden Markov Models (HMM) in predicting cardiovascular diseases. Problem: The addressed issue is the need to enhance the early detection of heart diseases, emphasizing how HMM can address uncertainty in clinical data and detect complex patterns. Objective: To evaluate the use of Hidden Markov Models (HMM) in the analysis of electrocardiograms (ECG) for the early detection of cardiovascular diseases. Methodology: The methodology comprises a literature review concerning the relationship between HMM and cardiovascular diseases, followed by the application of HMM to prevent heart attacks and address uncertainty in clinical data. Results: The findings indicate that HMM is effective in preventing heart diseases, yet its effectiveness is contingent upon data quality. These results are promising but not universally applicable. Conclusions: In summary, this study underscores the utility of HMM in early infarction detection and its statistical approach in medicine. It is emphasized that HMM is not infallible and should be complemented with other clinical options and assessment methods in real-world situations. Originality: This work stands out for its statistical and probabilistic approach in the application of Hidden Markov Models (HMM) in medical analysis, offering an innovative perspective and enhancing the understanding of their utility in the field of medicine. Limitations: It is recognized that there are limitations, such as dependence on data quality and variable applicability in clinical cases. These limitations should be considered in the context of their implementation in medical practice.