International Journal of Science, Engineering and TechnologyAuthors: Shail Sahu, Neelam Sahu, Harish Kumar
Abstract: This study introduces a predictive healthcare framework that applies Logistic Regression to assess individual health risks based on lifestyle and physiological indicators. The system incorporates variables such as age, body mass index (BMI), exercise frequency, dietary habits, smoking behavior, and prior medical records to estimate the probability of developing lifestyle-related illnesses. The model was trained and validated on a structured dataset, achieving a classification accuracy of 98.8%. Findings highlight that Logistic Regression, though relatively straightforward compared to more complex algorithms, delivers dependable and interpretable outcomes. Its transparency makes it particularly suitable for healthcare contexts, where understanding the influence of each factor is essential. The proposed approach has potential applications in early risk detection and preventive health planning, supporting clinicians and individuals in making informed decisions.
DOI: http://doi.org/10.5281/zenodo.20175475
