Data Mining for Heart Disease Prediction Based on Echocardiogram and Electrocardiogram Data

Traditional methods of detecting cardiac illness are often problematic in the medical field. The doctor must next study and interpret findings patient's record received from electrocardiogram echocardiogram. These tasks take a long time require patience. use computational technology medicine, especially disease, is not new. Scientists continuously striving for most reliable method diagnosing illness, particularly when an integrated system constructed. attempted to propose alternative identifying using supervised learning technique, namely multi-layer perceptron (MLP). started with collection patient data, which yielded up 534 data points, followed by pre-processing transformation provide 324 points suitable be employed algorithms. last step create heart disease classification model distinct activation functions MLP. degree accuracy, k-fold cross-validation, bootstrap all used test model. According study, MLP Tanh function more accurate prediction than logistics Relu. accuracy level (CA) cross-validation 0.788 data-sharing situation, while it 0.672 Bootstrap. best based on CA AUC value, values 0.832 (k-fold cross-validation) 0.857 (bootstrap).