Prediction of Heart Disease Using Hybrid of CNN- LSTM Algorithm

Abstract

Cardiovascular illness is the leading cause of death globally. Early detection and faultless prediction of heart disease can significantly improve patient outcomes. Thus, the hybrid of CNN-LSTM (Convolutional Neural Network – Long Short-Term Memory) algorithm is employed for determining the likelihood of heart illness. The algorithm is trained and tested using the UCI repository dataset, which contains various features such as age, sex, cholesterol levels, blood pressure, etc. The proposed algorithm first incorporates a Convolutional Neural Network (CNN) to obtain relevant characteristics derived from the input data. Then, the output of CNN is fed into a long short-term memory (LSTM) network, which is capable of capturing the temporal dependencies in the data. The resulting model is capable of predicting the occurrence of heart disease with high accuracy. Python programming language is used to implement the proposed algorithm, making use of the TensorFlow and Keras libraries for building and training the model. The dataset is processed to deal with missing values and normalize input features. Further model's performance is assessed using several metrics, including accuracy, precision, recall and F1-score. The result demonstrates that the proposed CNN-LSTM hybrid model outperforms traditional machine learning algorithms and other neural network architectures in predicting heart disease.