# H-DSAE: a hybrid technique to recognize heart disease

## TL;DR

This paper introduces H-DSAE, a hybrid machine learning technique that improves heart disease diagnosis with high accuracy using deep learning and ensemble methods.

## Contribution

The novel H-DSAE method combines DBN, SVM, and SAE for heart disease classification with high diagnostic accuracy.

## Key findings

- H-DSAE achieved 99.2% accuracy in heart disease classification.
- The method demonstrated 97.5% sensitivity and 98.5% F-measure.
- It outperforms conventional diagnostic methods by reducing human error and improving prediction speed.

## Abstract

Over the years, the number of people who succumbed to heart ailments has increased significantly worldwide. The World Health Organization claims that about 17 million people die each year due to heart disease. High levels of cholesterol and blood pressure are some risk factors. This technology seeks to treat these conditions before they become a problem. Through machine learning, doctors can now make more informed decisions regarding the treatment of patients. Machine learning can assist in reducing the likelihood of a cardiac event. Conventional methods for diagnosing diseases often lead to inaccurate diagnoses and take longer to complete due to human errors. In order to increase the diagnostic accuracy, an ensemble method is used. This method combines various classifiers to achieve highly accurate predictions. Due to the complexity of the task, the researchers decided to use deep learning methods to perform the heart disease classification task. H-DSAE technique utilize Deep Belief Network (DBN), Support Vector Machine (SVM), and Stacked Auto-Encoder (SAE). It was able to extract various heart image representations and achieve an accuracy of 99.2. It also had a sensitivity of 97.5, F-measure of 98.5, and precision of 98.4. The next phase of the project will focus on developing more advanced classification and features algorithms. This will help improve the efficiency of the system.