Machine Learning Algorithms to Detect Heart Diseases


Avishek Sharma
Techno College of Engineering Agartala, Maheshkhala, Tripura
Pranta Sutradhar
Techno College of Engineering Agartala, Maheshkhala, Tripura
Parijata Majumdar
Techno College of Engineering Agartala, Maheshkhala, Tripura


Background: The heart is one of our body's most essential organs, pumping blood to various organs via veins, arteries. Heart disease is a disorder that impairs the heart's ability to operate. It has now evolved into a serious illness that shortens human life.

Objectives: A difficult challenge in the healthcare industry is forecasting cardiac disease data which refers to a large volume of data collected at a rapid pace. These algorithmic approaches [1] will be implemented, executed on the dataset to give best results so that medical practitioners can make rapid 'decisions and diagnose accurately.

Methodology: Algorithms Used: The algorithms used are Perceptron, Decision Tree Classifier [2], Random Forest, and K-Neighbors [3].

Libraries used: Numpy: It is the most important Python module for scientific computing and mathematical computations. Scikit-learn: -Sk learn in Python, Scikit-learn (Sklearn) is the most usable and robust machine learning package. Pandas: It is a data manipulation and analysis software package for the Python programming language. Matplotlib: It is used for plotting graphs from numeric data in python.

Results and Discussion: If two supervised learning algorithms were used in separate experiments, there's a risk that a performance comparison between them might provide error results. These studies employed a variety of factors or measurements to predict illness [4].

Conclusion and future scope: Due to the scarcity of symptom and diagnostic data, data processing and validation of heart-related diseases is a difficult task. The future scope involves using a hybrid ML algorithm utilizing Decision Tree's features with X-ray image classification system for better prediction of results [5].

January 28, 2022