Improving Heart Disease Prediction Accuracy Using Evolutionary Algorithms and CNN Models

Abstract

Heart disease prediction is critical in the healthcare system due to the disease's high-risk factor. Data analysis is critical in predicting outcomes depending on medical history. Each factor must be taken into account to ensure that the prediction is accurate. Conventional methods rely on huge amounts of information rather than precise prediction. Data must be chosen carefully in order to achieve an earlier predictive process. If data collection is incomplete, analysis is harmed. As a result, the work proposed improving prediction accuracy by transforming missing information to indeed from the dataset. The method incorporates a CNN-MMEI classifier for previous accurate prediction (efficient multi - modality disease early identification utilizing convolutional neural networks (CNNs)) with an approach. Using data sets, the Nave Bayes algorithm is used to process the dataset. The suggested neuro-genetic approach identifies a feasible configuration for the optimal network. The results demonstrate that by combining an effective algorithm and a classifier, 85% percent accuracy is achieved. The performance indicators proposed will shed light on reliable variables.