| Title |
Automated hypertension detection using convolutional neural networks / |
| Authors |
Trakšelis, Kristupas ; Bikulčienė, Liepa ; Butkevičiūtė, Eglė |
| DOI |
10.15388/DAMSS.14.2023 |
| ISBN |
9786090709856 |
| Full Text |
|
| Is Part of |
DAMSS 2023:14th conference on data analysis methods for software systems, November 30 – December 2, 2023, Druskininkai, Lithuania.. Vilnius : Vilnius University press, 2023. p. 91.. ISBN 9786090709856 |
| Keywords [eng] |
ECG signal analysis ; convolutional neural networks ; hypertension recognition ; heartbeat classification |
| Abstract [eng] |
Cardiovascular diseases, notably hypertension, pose a significant threat to global public health, contributing substantially to mortality rates. Timely diagnosis and intervention are crucial in preventing the adverse consequences of heart disorders, including damage to vital organs. The primary objective of this study is to develop a robust ECG-based methodology for heart disease detection and classification, with a specific focus on distinguishing between hypertensive and healthy patients. The ECG signals were carefully prepared by removing noise, and correcting baseline wander. Furthermore, the heartbeats were automatically identified and isolated from the ECG signal data. These segmented heartbeats were plotted on a 480x480px image and classified using Convolutional Neural Network (CNN). In this study, the open source SHAREE and PTB databases were utilized, which include 139 hypertensive and 52 healthy patients respectively. Created model was evaluated on real-world data taken from hypertensive and healthy patients. Our model successfully classified patients with higher than 96% accuracy. This study proposes a novel ECG-based methodology for detecting hypertensive patients. It shows that it is easily implementable and capable of classifying with high accuracy. |
| Published |
Vilnius : Vilnius University press, 2023 |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2023 |
| CC license |
|
