An optimized deep belief system for heart disease classification and severity prediction

Artificial Intelligence (AI) is applicable in many digital applications such as education, medical, transactions, etc.; it has afforded the finest results in all application sectors. Besides, smartly analyzing diseases is required in today's life scenario. However, the vast and unstructured data has complicated the disease specification. So, the present study is interested in designing a novel Chimp-based Deep Belief Model (CbDBM) for forecasting heart failure and arrhythmia. The dataset for this current study is heart electrocardiogram (ECG) numerical data. Initially, the noise contents in the data are filtered at the preprocessing stage. Moreover, based on the fitness process of the chimp, efficient features were extracted, and the data were classified as normal and abnormal. This model is tested in Python, and the results are validated. The model acquired 97.4% accuracy, recall, precision and f-score, which are higher than the traditional models. Hence, the system is effective for heart disease prediction.