Reliable AI-Powered ECG Anomaly Detection and Decision Support System for Instantaneous Medical Monitoring
ID：31
Submission ID：262 View Protection：ATTENDEE
Updated Time：2026-07-25 18:01:41
Hits：29
Online
Start Time：2026-07-30 16:50 （Asia/Kolkata）
Duration：15min
Session：[S4] Computer Vision and Pattern Recognition » [S4-3] Computer Vision and Pattern Recognition
Video
No Permission
Presentation File
Attachment File
Tips: The file permissions under this presentation are only for participants. You have not logged in yet and cannot view it temporarily.
Abstract
Primarily in remote and resource-constrained areas, accurate and timely healthcare tracking has become crucial for the early identification of cardiac problems. The Reliable equipped with artificial intelligence ECG anomaly detection and decision support framework presented in this paper integrates XGBoost, Artificial Neural Network (ANN), simulated vital monitoring, and real-time interaction technologies. Before extracting significant statistical and morphological properties such rolling mean, variability, slope, minimum, and maximum values, ECG data from the MIT-BIH Arrhythmia Database are preprocessed to eliminate baseline drift and high-frequency noise. To increase the resilience and accuracy of classification, a hybrid ANN–XGBoost model is created. The suggested method detects ECG anomalies with 98.6% accuracy, 98.1% precision, 97.8% recall, and a 97.9% F1-score. A real-time visualization dashboard based on Streamlit and a WhatsApp alert system that uses the Twilio API to enable emergency notifications in three to five seconds are also included in the framework. The findings of the experiment show enhanced interpretability, dependability, and real-time health monitoring capabilities.
Keywords
ECG anomaly detection, Reliable AI, ANN, XGBoost, IoMT, Decision Support System, Real-time healthcare monitoring, AI-enabled networks
Speaker
Comment submit