A Review on Vision Transformer and Explainable AI Approaches for ECG-based Heart Disease Detection

Heart diseases continue to be major cause of death in the world, World Health Organization estimates that 17.9 million people die each year from cardiovascular illnesses. So, there is a need for accurate diagnosis to reduce the death rates. Electrodiagram(ECG) signals play an important role in identifying heart abnormalities. Traditionally , ECG signals are used to detect the heart diseases using convolutional neural networks (CNN) and Machine learning .In order to categorize ECG signals, CNN is employed, These methods mainly focus on the individual waveform patterns ,missing the broader context of ECGs as seen by doctors. This is the main disadvantage which reduces the system ability to fully understand the overall heart activity. To overcome this , A review on existing techniques is performed and proposed an approach that uses ECG images rather than ECG 1D signals. By implementing Vision Transformers(ViT),a deep learning architecture which treats the ECGs as complete images so that the model is able to find temporal and global relationships present in the image. This helps the model to see ECGs same as how the cardialogists see them. In addition to this, explainability tools such as heatmaps and attention maps are used to highlight the important regions in the ECGs that influence the model to take that prediction. This promotes interpretability which helps to gain doctor’s trust. This approach bridges the gap between research and real world implementation

• Publication year

  2025

• Venue

  2025 5th International Conference on Evolutionary Computing and Mobile Sustainable Networks (ICECMSN)

• Publication date

  2025-11-24

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/ICECMSN68058.2025.11382889
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• ECGNet: High-Precision ECG Classification Using Deep Learning and Advanced Activation Functions

  2025cited by this paper
• GMAR: Gradient-Driven Multi-Head Attention Rollout for Vision Transformer Interpretability

  2025cited by this paper
• Comparative Study of Vision Transformer (ViT), Swin Transformer, and Transformer in Transformer (TNT) on Image Classification

  2025cited by this paper
• Prediction and analysis of cardiovascular diseases based on ECG images using computer vision algorithms

  2025cited by this paper
• Selective Weighting and Prediction Error Expansion for High-Fidelity Images

  2024cited by this paper
• Pixel Value Ordering Based Prediction for High-Fidelity Images

  2024cited by this paper
• Vision Transformers for Biomedical Applications *

  2024cited by this paper
• Heart Disease Detection Using Machine Learning

  2024cited by this paper
• Optimizing Heart Disease Prediction with Random Forest: Insights from the Kaggle Dataset

  2024cited by this paper
• Explainable AI-Based ECG Heartbeat Classification Using Deep Learning Models

  2024cited by this paper
• Corrections to "Comparative Analysis of Machine Learning Algorithms With Advanced Feature Extraction for ECG Signal Classification"

  2024cited by this paper
• Heart Disease Prediction Using Enhanced Whale Optimization Algorithm Based Feature Selection With Machine Learning Techniques

  2023cited by this paper
• Deep Learning and Vision Transformer for Medical Image Analysis

  2023cited by this paper
• Accurate Gradient Selective Based Histogram Bin Shifting for Reversible Data Hiding

  2023cited by this paper
• Gradient Directional Predictor for Reconstructing High-Fidelity Images

  2023cited by this paper
• Gradient Adaptive Planar Prediction for High-Fidelity Images

  2023cited by this paper
• Vision Transformer (ViT)-based Applications in Image Classification

  2023influential reference
• Analyzing Clinical 12-Lead ECG Images Using Deep Learning Algorithms for Objective Detection of Cardiac Diseases

  2022cited by this paper
• Classification of heart disease from ECG signals using Machine Learning

  2021cited by this paper
• A Comparative Analysis of Deep Neural Network Models using Transfer Learning for Electrocardiogram Signal Classification

  2021cited by this paper
• 3-D ECG images with Deep Learning Approach for Identification of Cardiac Abnormalities from a Variable Number of Leads

  2021cited by this paper
• Heart Disease Prediction using Supervised Machine Learning Algorithms

  2020cited by this paper
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

  2020cited by this paper
• Methodologies and Techniques for Heart Disease Classification and Prediction

  2020cited by this paper
• Reversible Data Hiding with Selected Directional Context Based Prediction Using 8-Neighborhood

  2020cited by this paper
• Signal to Image to Classification: Transfer Learning for ECG

  2020cited by this paper
• Reversible Data Hiding Using Multi-Layer Perceptron Based Pixel Prediction

  2020cited by this paper
• Reversible Data Hiding Using Improved Gradient Based Prediction and Adaptive Histogram Bin Shifting

  2020cited by this paper
• ECG Heartbeat Classification Using Convolutional Neural Networks

  2020cited by this paper
• Beginnings of Artificial Intelligence in Medicine (AIM): Computational Artifice Assisting Scientific Inquiry and Clinical Art – with Reflections on Present AIM Challenges

  2019cited by this paper
• Heart Disease Prediction using Machine Learning

  2019cited by this paper
• Heart Disease Detection Using Machine Learning Majority Voting Ensemble Method

  2019cited by this paper
• Reversible Data Hiding Based on the Random Distribution of Reference Pixels

  2018cited by this paper
• IMPORTANCE OF COMPUTER VISION FOR HUMAN LIFE

  2017cited by this paper
• Attention is All you Need

  2017cited by this paper
• A Decomposable Attention Model for Natural Language Inference

  2016cited by this paper
• Selected Context Dependent Prediction for Reversible Watermarking with Optimal Embedding

  2016cited by this paper
• Gradient dependent reversible watermarking with low embedding impact

  2016cited by this paper
• Neural Machine Translation by Jointly Learning to Align and Translate

  2014cited by this paper

• No citing papers are available for this paper.