Litcius/Paper detail

Random Forest for Heart Disease Detection: A Classification Approach

Sanjay Dhanka, Surita Maini

20212021 IEEE 2nd International Conference On Electrical Power and Energy Systems (ICEPES)47 citationsDOI

Abstract

Heart disease (HD) diagnosis is one of the most crucial task in the field of medical diagnosis. HD detection/prediction in early stages can save millions of life. Machine Learning (ML) is playing a substantial role in HD detection. Cardiologist, Physicians and Biomedical Engineers has collectively designed different ML algorithms to detect HD in initial stages. In this paper, authors proposed a Random Forest (RF) ML algorithm to predict the HD based on general physiological parameters. The RF algorithm is applied on Cleveland heart disease dataset having the record of 303 patient with 14 physiological attributes. This dataset is splitted in 80:20 ratio; 80% for training purpose and 20% for testing of RF model. Performance of the model was validated by 10-Fold cross validation and the performance metrics was analyzed. The proposed RF algorithm claimed the accuracy of 90.16%.

Topics & Concepts

Random forestComputer scienceTask (project management)Machine learningArtificial intelligenceStatistical classificationField (mathematics)Heart diseaseMedicineEngineeringInternal medicineMathematicsSystems engineeringPure mathematicsArtificial Intelligence in HealthcareMachine Learning in HealthcareImbalanced Data Classification Techniques
Random Forest for Heart Disease Detection: A Classification Approach | Litcius