Introduction

Convolutional neural networks (CNNs) can be used to classify electrocardiogram (ECG) beats in the diagnosis of cardiovascular disease.

ECG signals are typically processed as one-dimensional signals while CNNs are better suited to multidimensional pattern or image recognition applications.

In this Project, the morphology and rhythm of heartbeats are fused into a two-dimensional information vector for subsequent processing by CNNs that include adaptive learning rate and biased dropout methods.

It has been reported that about 80% of sudden cardiac deaths are the result of ventricular arrhythmias or irregular heartbeats.

While an experienced cardiologist can easily distinguish arrhythmias by visually referencing the morphological pattern of the ECG signals, a computer-oriented approach can effectively reduce the diagnostic time and would enable the e-home health monitoring of cardiovascular disease.

To achieve this we proposed CNN for automation of diagnosing ECG disease to notify the patient as well as doctor on time.

Motivation

It has been reported that about 80% of sudden cardiac deaths are the result of ventricular arrhythmias or irregular heartbeats.

They don’t get on time treatment or first aid

Understanding ECG pattern is even more complex task

The proposed solution will easily classify ECG diseases as well as notify patient and Doctor.

Objectives

The main goal of the project is to classify ECG related disease through CNN.

This include sub objective which are as follows:

  • Preparing data and its annotations
  • Since CNN is working on 2D therefore we will convert 1D to 2D.
  • Tuning the training parameters such as epochs and dataset.
  • Training Data on CNN model.
  • Testing data on CNN model to check efficiency
  • Applying real time signal processing and disease detection

Group Members

  • Malik Muhammad Hamza     

  • Abdullah

  • Muhammad Adnan 

 

Software Desingn

The Software design has four major parts major parts:

1.Data and segmentation

2.Training and Testing

3.Simulation and real time detection

4.Python code for Convolutional neural network

Implementation of Design

The project is using Anaconda repository for downloading required packages

The project is coded purely in python 3.7 language

The project is implemented in jupyter notebook IDE

The project is using Convolutional Neural Network paradigm

Simulation

Commercialization

Commercial application

1.Diagnosing disease

2.Health Technology

3.Health Laboratories 

4.Health Researches

Target Beneficiaries

1.E-Health

2.Sportsman 

3.Heart patient 

4.Aged persons 

Over 300 + Alumni Working in leading national & international firms

Including  TCON Constructions, UAE, GEM – UAE, Daud Sons, Bacha Khan International Airport