Publication

Abstract : Heart disease has seriously threatened people's health in recent decades due to its prevalence and high mortality rate. Detecting heart disease through clinical features is a major challenge in today’s world. Machine Learning (ML) and Deep Learning (DL) are technological innovations now being effectively used in healthcare, disease prediction, and biomedical care. This paper proposes a model RL-DLBH for identifying the Risk Levels (RL), to select the best features using BHHO(BH) and a deep learning model based on a Deep bi-LSTM (DL) classifier for detecting heart diseases and determining their risk level. The method was created by first choosing the most important attributes from the dataset by using BHHO to determine whether the patient has heart disease and the level of heart disease risk in patients based on their clinical report, and then using a deep Bi-LSTM model as a classifier. The characteristics of heart disease are defined by the main risk factors. ST depression, the highest heart rate, cholesterol, and chest pain are all factors to consider. Class labels were assigned to the following risk levels: risk level 1, risk level 2, and risk level 3. Three distinct datasets namely Cleveland, Hungarian, and CH (Cleveland and Hungarian) are used in this work. The experimental results show that BHHO with Deep bi-LSTM performs well with a classification accuracy of 98.12% compared with the existing models.