Prediction and Classification of Electrocardiogram-Signals using Machine Learning using Apache Spark

Gauravaram, Rishab Rao (2021) Prediction and Classification of Electrocardiogram-Signals using Machine Learning using Apache Spark. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (4MB) | Preview

Abstract

The Electrocardiogram (ECG) is a common diagnostic system to identify cardiovascular diseases (CDVs). The aim of this research project is to develop data mining models using Support Vector Machines, Random Forest and Deep Neural Networks in Apache Spark that can be applied on large ECG data to extract and classify signals and predict cardiac Arrhythmia. This will help medical practitioners’ timely diagnosis of diseases and help provide timely treatment to the patients. This research paper mainly focuses on classifying the ECG signal into five different classes of signals in the MIT-BIH dataset using Apache Spark. The performance of the models will be evaluated using Accuracy, Specificity, Sensitivity, and computation time. The Deep Neural Network model (local implementation) achieved a classification accuracy of 98.69% (sensitivity 97.01% and Specificity 99.29%). The Deep Neural Network model on Apache Spark achieved an accuracy of 93.94% (sensitivity 53.14%, specificity 93.89%). Overall, all the models created achieved high accuracy and specificity.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software R Medicine > Healthcare Industry
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Clara Chan
Date Deposited:	29 Nov 2021 12:01
Last Modified:	29 Nov 2021 12:01
URI:	https://norma.ncirl.ie/id/eprint/5152

Actions (login required)

View Item