Identifying Heart Attack Risk in Vulnerable Population: A Machine Learning Approach

The COVID-19 pandemic has significantly increased the incidence of post-infection cardiovascular events, particularly myocardial infarction, in individuals over 40. While the underlying mechanisms remain elusive, this study employs a hybrid machine learning approach to analyze epidemiological data in assessing 13 key heart attack risk factors and their susceptibility. Based on a unique dataset that combines demographic, biochemical, ECG, and thallium stress-tests, this study categorizes distinct subpopulations against varying risk profiles and then divides the population into 'at-risk' (AR) and 'not-at-risk' (NAR) groups using clustering algorithms. The study reveals strong association between the likelihood of experiencing a heart attack on the 13 risk factors studied. The aggravated risk for postmenopausal patients indicates compromised individual risk factors due to estrogen depletion that may be, further compromised by extraneous stress impacts, like anxiety and fear, aspects that have traditionally eluded data modeling predictions.

VIRDOC: Statistical and Machine Learning by a VIRtual DOCtor to Predict Dengue Fatality

Clinicians conduct routine diagnosis by scrutinizing signs and symptoms of patients in treating epidemics. This skill evolves through trial-and-error and improves with time. The success of the therapeutic regimen relies largely on the accuracy of interpretation of such sign-symptoms, based on which the clinician ranks the potent causes of the epidemic and analyzes their interdependence to devise sustainable containment strategies. This study proposed an alternative medical front, a VIRtual DOCtor (VIRDOC), that can self-consistently rank key contributors of an epidemic and also correctly identify the infection stage, using the language of statistical modelling and Machine Learning. VIRDOC analyzes medical data and then translates these into a vector comprising Multiple Linear Regression (MLR) coefficients to probabilistically predict scores that compare with clinical experience-based assessment. The VIRDOC algorithm, risk managed through ANOVA, has been tested on dengue epidemic data (N=100 with 11 weighted sign-symptoms). Results highly encouraging with ca 75% accurate fatality prediction, compared to 71.4% from traditional diagnosis. The algorithm can be generically extended to analyze other epidemic forms.