Key result
L2-regularized logistic regression classifier enables preliminary heart disease risk assessment on clinical data.
Why the study?
Early identification of heart disease risk enables timely medical advice, but manually analyzing multiple health-related factors can be difficult and error-prone.
Does an L2-regularized Logistic Regression model with an adjusted decision threshold accurately classify heart disease risk in patients from the UCI Heart Disease (Cleveland) dataset?
Observational
Does an L2-regularized Logistic Regression model with an adjusted decision threshold accurately classify heart disease risk in patients from the UCI Heart Disease (Cleveland) dataset?
This project demonstrates the use of an L2-regularized logistic regression machine learning model on the UCI Heart Disease (Cleveland) dataset for preliminary, interpretable heart disease risk assessment, though no quantitative results are reported in the abstract.
Answers are grounded in this paper and cite their sources.
May support recall-prioritized risk prediction from routine data; leaves open prospective validation before clinical adoption.
Heart disease is one of the major health problems affecting people across the world. Identifying the possibility of heart disease at an early stage can help in taking timely medical advice and reducing potential risks. However, analysing different health-related factors manually can sometimes be difficult and may lead to errors. To address this problem, this project introduces a Medical Risk Classifier, a machine learning-based system designed to predict the risk of heart disease using basic clinical information. The system uses the UCI Heart Disease (Cleveland) dataset, which contains patient details such as age, blood pressure, cholesterol level, chest pain type, maximum heart rate and other medical attributes. The collected data is cleaned and prepared before applying a regularized Logistic Regression algorithm to classify the possibility of heart disease. L2 regularization is used to reduce overfitting and improve the stability of the model. The system is evaluated using stratified 5-fold cross-validation, along with performance measures such as precision, recall, F1-score, ROC-AUC and confusion matrix. A key focus of this project is to improve the model's ability to identify patients who may have heart disease rather than depending only on overall accuracy. For this reason, the decision threshold is adjusted to give greater importance to recall and reduce the chances of missing positive cases. The project also examines the contribution of different clinical features to the model's predictions. Overall, the Medical Risk Classifier demonstrates how machine learning can be used to analyse medical data and support preliminary heart disease risk assessment through a simple and interpretable approach. It is developed for academic and experimental purposes and is not intended to replace professional medical diagnosis.
No takes yet. Share an insight, caveat, or question.
Nishika et al. (2026) conducted an observational in Heart disease. Medical Risk Classifier (L2-regularized Logistic Regression) was evaluated on Heart disease risk classification performance (precision, recall, F1-score, ROC-AUC). An L2-regularized logistic regression classifier evaluated with 5-fold cross-validation on the UCI Cleveland dataset supported preliminary heart disease risk assessment.
Synapse has enriched 10 related papers. Consider them for comparative context: