Heart disease prediction using neuro-genetic algorithm and CNN-MDRP classifier

Abstract

Prediction of heart disease is vital in healthcare sector due to high risk factor related to the disease. Data analysis plays a vital role in prediction based on patient history. Each factor has to be taken intoconsideration for the prediction to be accurate. Conventional methods involve enormous data rather than accurate prediction. Data has to be chosen correctly for attaining earlier predicting process. If the data collected is partial it’s a setback for analysis. The previous work designed an Improved Step Adjustment based Glowworm Swarm Optimization Algorithm with Weighted Feature based Support Vector Machine (ISAGSO-WFSVM) for Heart diseasediagnosis. However, the WFSVM is does not suitable for large data sets and it has long training time. To solve this problem, the proposed system designed a Neuro-Genetic with CNN-MDRP approach for heart disease prediction. In this work, initially perform cleaning of data by converting the incomplete data to informational data from the dataset. The work combines CNN-MDRP classifier for earlier accurate prediction (convolutional neural network (CNN)-based efficient multimodal disease risk prediction)along with neuro genetic approach.The dataset is processed by Naïve Bayes algorithm using structured data. The proposed neuro-genetic approach finds a feasible solution for optimal network configuration. The results prove that combining both effective algorithm and classifier acquires 96.25% of accuracy. The metrics proposed will provide a clear insight on reliable factors.