A Statistical Analysis for Heart Disease Prediction System for Next-Gen Software

Nowadays, heart-related diseases are common to people of all ages. Early detection of heart-related anomalies can help save a life. This generates a demand for an efficient heart disease prediction system. An individual can use this prediction system to track his heart-related irregularities and be deployed in the medical health field and hospitals. This paper proposes a learning algorithm based heart disease prediction system. The proposed work is implemented on two learning algorithms, logistic regression and random forest algorithms, based on overall classification performance. In the proposed paper, an evaluation tool can model classifier simulators for datasets of patients having ill heart characteristics and performance metrics such as positive rates and performance accuracy under specific conditions. The proposed system attained relatively different classification scores under different conditions of the simulated dataset while implementing logistic regression and random forest algorithms. This proposed system can be deployed in the medical field and hospitals for real-time heart patient monitoring. The proposed system projected an accuracy of85%with logistic regression and 80%with random forest.