CONVOLUTIONAL NEURAL NETWORK MODEL FOR HEART DISEASE CLASSIFICATION

Subject Area: BIOMETRIC TECHNOLOGY

Heart disease has dominated the cause of mortality rate in the world. In western African, high kolanut intake has been reported to be one of the main factors which have provoked the rising cases of the problem among the elderly within this global region. The aim of this research is to develop a classification model for heart disease considering kolanut as a risk factor. This was achieved using Magnetic Resonance Image (MRI) data collection from the Enugu State University Teaching Hospital, Parklane, these data were augmented and then used to train a convolutional neural network algorithm to generate a classification model. The result of the model was evaluated considering accuracy and loss function which reported 99.01% and 0.011241 after tenfold validation. The model was compared with other state of the art algorithm and then result reported that the new system was more reliable when compared to the rest.