Scribd
Upload
10 views1 page

Project

The document outlines a project for implementing an intelligent edge computing framework for real-time cardiac arrhythmia classification using the TMS320C6748 DSP. It details the methodology for ECG signal acquisition, preprocessing, feature extraction, and deployment of a lightweight machine learning model for classification. The expected outcomes include low-latency, privacy-preserving cardiac monitoring suitable for wearable and remote healthcare applications.

Uploaded by

rajkumarmeena0711
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views1 page

Project

The document outlines a project for implementing an intelligent edge computing framework for real-time cardiac arrhythmia classification using the TMS320C6748 DSP. It details the methodology for ECG signal acquisition, preprocessing, feature extraction, and deployment of a lightweight machine learning model for classification. The expected outcomes include low-latency, privacy-preserving cardiac monitoring suitable for wearable and remote healthcare applications.

Uploaded by

rajkumarmeena0711
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Maulana Azad National Institute of Technology, Bhopal

Implementation of an Intelligent Edge Computing Framework for Real-Time Cardiac


Arrhythmia Classification on TMS320C6748 DSP

Introduction
Cardiovascular diseases are among the leading causes of mortality worldwide. Early detection of
arrhythmias through continuous ECG monitoring can significantly reduce life-threatening risks.
Conventional systems rely on cloud-based processing, which introduces latency, increased bandwidth
usage, and data privacy concerns. This project proposes an Edge AI-based real-time ECG arrhythmia
detection system implemented on the EPB-C6748, built around the TMS320C6748 from Texas
Instruments.

Proposed Methodology
ECG signals will be acquired using an analog front-end module and digitized for processing on the DSP
platform. Signal preprocessing techniques such as bandpass filtering, baseline wander removal, and noise
suppression will be implemented to enhance signal quality. Key features including QRS complex detection,
RR interval measurement, and heart rate variability will be extracted using optimized DSP algorithms.
A lightweight machine learning model (e.g., Support Vector Machine or Convolutional Neural Network)
will be trained offline using standard datasets such as the MIT-BIH Arrhythmia Database. The optimized
model will then be deployed onto the EPB-C6748 board for real-time edge inference.

Expected Outcomes
The system will classify ECG signals into Normal rhythm, Bradycardia, Tachycardia, and Atrial
Fibrillation. Performance will be evaluated based on accuracy, processing latency, memory utilization, and
power efficiency. The proposed solution offers low-latency, privacy-preserving, and scalable cardiac
monitoring suitable for wearable and remote healthcare applications.

Group Members Signature:


Supervisor Signature:
Dr. Punjan Dohare 1. Rajkumar Meena(2311301162)

2. Rounit Raj (2311301105)

3. Karan Prajapat (2311301151)

You might also like