Multimodal deep learning for automated 12-lead ECG diagnosis | Macro-AUC 0.927 on PTB-XL | InceptionTime1D + SE Attention + PTB-XL+ Feature Fusion | Edge AI (INT8/ONNX) | Interactive Web Dashboard

Code repository of the paper: Beyond Supervision: Evaluating Contrastive Self-Supervised Learning Techniques for Electrocardiogram-Based Mental Stress Detection

A medical AI desktop app featuring a custom-trained Deep Learning model and a self-curated dataset for high-accuracy ECG abnormality detection.

ML-based solution for ECG signal processing and diagnosis classification. Paper in SSPP 2025

Python port of ecgScorer https://github.com/ecgScorer/ecgScorer

Code repository of the paper: Electrocardiogram-Based Mental Stress Detection Amid Everyday Activities Using Machine Learning: Model Development and Validation Study. Published in the Journal of Medical Internet Research.

This is the official repository for CardioLab. A machine and deep learning framework for the estimation and monitoring of laboratory abnormalities throught ECG data.

Dual-channel CNN-LSTM for ECG arrhythmia classification (NSR, PAC, PVC) using raw ECG and derivative signals with Docker support

Deep learning system for automatic cardiac arrhythmia classification from 12-lead ECG signals. Implements CNN, LSTM, and hybrid architectures trained on PTB-XL dataset. Features multi-label classification for 5 diagnostic classes (NORM, MI, STTC, CD, HYP) with comprehensive visualization tools.

BioDG is a publically available framework for the evaluation of Domain Generalization algorithms in Biosignal Classification.

An ECG Monitoring System with Real-time Analysis for Tele-medecine facilites

Official Repo for, GRACE: Privacy-centric Multimodal Seizure Detection

Demo of a smartwatch based systematic health monitoring solution designed for patients with chronic conditions

A Method to Improve Any ECG Denoising Technique In limb leads

Oloche's AI Cardiologist is a deep learning web app for real-time automated classification of cardiac arrhythmias from raw ECG signals. Uses a custom 1D CNN trained on MIT-BIH database to classify heartbeats into five categories with confidence scores and visualizations for diagnostic support.

A deep learning-based system for automatic detection of sleep apnea from ECG signals using a hybrid 1D CNN-BiLSTM architecture with an attention mechanism. Achieves high accuracy with minimal preprocessing, making it suitable for real-time, portable diagnostic applications.

Evaluation of Deep Learning models for detecting irregular heartbeat rhythms (arrhythmias) on electrocardiogram (ECG) measurements.

Clasificación de señales de Electrocardiogramas (ECG) mediante Deep Learning. Implementación basada y entrenada con el dataset MIT-BIH. Incluye una aplicación web interactiva con Flask.

Repository for the paper 'Prospects for AI-Enhanced ECG as a Unified Screening Tool for Cardiac and Non-Cardiac Conditions -- An Explorative Study in Emergency Care'.

[Biomedical Signal Processing and Control] ECGTransForm: Empowering adaptive ECG arrhythmia classification framework with bidirectional transformer