Early Heart Attack Detection Using Hybrid Deep Learning Techniques

Given the significant risk that heart disease, particularly heart attacks, poses to individuals’ lives, it is crucial to develop effective techniques for early detection. Advanced machine learning and deep learning algorithms have the ability to predict heart attacks by analyzing a patient’s medical...

Full description

Saved in:
Bibliographic Details
Main Authors: Niga Amanj Hussain, Aree Ali Mohammed
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Information
Subjects:
Online Access:https://www.mdpi.com/2078-2489/16/5/334
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711276769411072
author Niga Amanj Hussain
Aree Ali Mohammed
author_facet Niga Amanj Hussain
Aree Ali Mohammed
author_sort Niga Amanj Hussain
collection DOAJ
description Given the significant risk that heart disease, particularly heart attacks, poses to individuals’ lives, it is crucial to develop effective techniques for early detection. Advanced machine learning and deep learning algorithms have the ability to predict heart attacks by analyzing a patient’s medical history and overall health. These algorithms can process large datasets, extracting valuable insights that help mitigate the risk of fatal outcomes. This study integrates a deep learning approach to predict and detect heart attacks early by classifying patient data as normal or abnormal. The proposed model combines a Convolutional Neural Network (CNN) with self-attention, leveraging the self-attention mechanism to focus on the most critical aspects of the sequence. Since heart attack risk is closely tied to the changes in vital signs over time, this approach enables the model to learn and assign appropriate weights to each input component. Improvements and modifications to the hybrid model resulted in a 98.71% accuracy rate during testing. The model’s strong performance on evaluation metrics shows its potential effectiveness in detecting heart attacks.
format Article
id doaj-art-7fde5baa89d1408eb776269d9c06f144
institution DOAJ
issn 2078-2489
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Information
spelling doaj-art-7fde5baa89d1408eb776269d9c06f1442025-08-20T03:14:39ZengMDPI AGInformation2078-24892025-04-0116533410.3390/info16050334Early Heart Attack Detection Using Hybrid Deep Learning TechniquesNiga Amanj Hussain0Aree Ali Mohammed1Computer Science Department, College of Science, University of Sulaimani, Sulaymaniyah 46001, IraqComputer Science Department, College of Science, University of Sulaimani, Sulaymaniyah 46001, IraqGiven the significant risk that heart disease, particularly heart attacks, poses to individuals’ lives, it is crucial to develop effective techniques for early detection. Advanced machine learning and deep learning algorithms have the ability to predict heart attacks by analyzing a patient’s medical history and overall health. These algorithms can process large datasets, extracting valuable insights that help mitigate the risk of fatal outcomes. This study integrates a deep learning approach to predict and detect heart attacks early by classifying patient data as normal or abnormal. The proposed model combines a Convolutional Neural Network (CNN) with self-attention, leveraging the self-attention mechanism to focus on the most critical aspects of the sequence. Since heart attack risk is closely tied to the changes in vital signs over time, this approach enables the model to learn and assign appropriate weights to each input component. Improvements and modifications to the hybrid model resulted in a 98.71% accuracy rate during testing. The model’s strong performance on evaluation metrics shows its potential effectiveness in detecting heart attacks.https://www.mdpi.com/2078-2489/16/5/334heart attack predictionhybrid CNN–self attention modeldeep learning in healthcaretime series dataabnormality detection
spellingShingle Niga Amanj Hussain
Aree Ali Mohammed
Early Heart Attack Detection Using Hybrid Deep Learning Techniques
Information
heart attack prediction
hybrid CNN–self attention model
deep learning in healthcare
time series data
abnormality detection
title Early Heart Attack Detection Using Hybrid Deep Learning Techniques
title_full Early Heart Attack Detection Using Hybrid Deep Learning Techniques
title_fullStr Early Heart Attack Detection Using Hybrid Deep Learning Techniques
title_full_unstemmed Early Heart Attack Detection Using Hybrid Deep Learning Techniques
title_short Early Heart Attack Detection Using Hybrid Deep Learning Techniques
title_sort early heart attack detection using hybrid deep learning techniques
topic heart attack prediction
hybrid CNN–self attention model
deep learning in healthcare
time series data
abnormality detection
url https://www.mdpi.com/2078-2489/16/5/334
work_keys_str_mv AT nigaamanjhussain earlyheartattackdetectionusinghybriddeeplearningtechniques
AT areealimohammed earlyheartattackdetectionusinghybriddeeplearningtechniques