Smart ECG

Smart ECG (Electrocardiogram) will be capable of detecting and analyzing patient?s ECG through a wireless sensor network. Data will be transmited to Cloud for diagnosis through WiFi technology. Wireless technology is not only convenient and reliable but is also necessary in today?s se

Project Title

Smart ECG

Project Area of Specialization

Biomedical Engineering

Project Summary

Smart ECG (Electrocardiogram) will be capable of detecting and analyzing patient’s ECG through a wireless sensor network. Data will be transmited to Cloud for diagnosis through WiFi technology. Wireless technology is not only convenient and reliable but is also necessary in today’s seamless internet-oriented world. It will assist cardiologists: they will be able to monitor the ECG of the patient from a remote location. It will have a sensing system. Sensing system will contain a network of ECG sensors for detecting patient’s heart activity in real time. The ECG recorded will then be sent to the cloud using a sensor network of WIFI technology where a novel machine leaning model will classify the signals as either normal or abnormal ones. Once any abnormality is detected, the system will then propose a treatment according to the type of abnormality and possibly recommend a doctor based on results.

Project Objectives

Pakistan is a developing country with a population of over 220 million and the main cause of deaths is due to cardiovascular diseases. The ratio of deaths is about 8% of the total deaths. Pakistan does not have the resources to combat this issue, most of the population lives in rural areas where non-availability of doctors and necessary facilities are the major causes of the high death rate.

In order to combat this serious threat, we propose to develop a working prototype of “Smart ECG” which is an automatic ECG monitoring and diagnostic device capable of detecting and analyzing ECG and proposing treatment according to the results. The ECG analysis system will be based on machine learning classifiers that will sample ECG signals of the patient’s heart and classify them as either normal or abnormal ones. There are 37 different abnormal ECG's and the model will be able to differentiate between each of them.

The main objectives of this project are;

• To provide accurate detection and analysis of a patient’s ECG in real-time.
• To provide real-time monitoring of patient’s heart activity in rural areas over a telecommunication infrastructure (IoT).
• To provide a reliable, cost-effective, practical and efficient way to diagnose ECG of the patient and prevent any future mishaps.
• To provide quality of care and ease of access to the patients.
• To provide improved disease management and improved outcomes of treatment.
• To provide enhanced patient experience.

Project Implementation Method

The proposed Smart ECG to detect, monitor and diagnose electrical activity of a patient’s heart is based on the following components.

• Sensing System
• Visualizing Component
• Central System
• Cloud Computing
• Result
• Display

1) Sensing System: This component will be responsible for detecting and capturing ECG signals of the patient’s heart. The sensing system consists of a low power, single lead, heart rate monitoring sensor designed to extract, amplify and filter ECG signals in the presence of noisy conditions such as those created by motion or remote electrode placement. The purpose of using a single lead sensor is portability and low power consumption.

2) Visualizing Component: This component will be responsible for displaying the ECG recorded from the above-mentioned system.

3) Central System: This component will send the recorded ECG to the cloud for analysis and diagnosis through WIFI sensor. The central system is where all the sensors are connected including ECG sensor, WIFI sensor, LCD Display etc.

4) Cloud Computing: The central system will send the ECG signals to cloud for analysis and diagnosis which have a machine learning model running in the back-end to classify whether the signal has any abnormalities or is a normal ECG signal, if the signal has any abnormality the model will be able to predict the type of abnormality based on the trained data and will propose treatment accordingly. The proposed model will be trained on the existing publicly available data of ECG recordings to separate the normal and abnormal signals. When the performance of the Machine learning algorithm comes in the acceptable range then it will go through a testing phase using local real-time testing data. The output of the Machine learning model will decide the result of the ECG recording.

5) Result: The output of Machine learning model will decide the result of the ECG recording. Once any abnormality is detected the model will then propose treatment according to the type of abnormality. There are total 37 different abnormal ECG types and each has its own treatment.

6) Display: Results along with the proposed treatment in case of abnormality will then be displayed on the LCD screen.

Benefits of the Project

This project will provide a unique and easy-to-use device for diagnosing ECG and will serve as an assisting tool for cardiologists all over the Pakistan. This device will enhance the overall diagnosing process of the ECG. Moreover distance and travel time between the patients living in the rural areas and the care providers will be eliminated as the device is designed to monitor and diagnose the patients heart activity over telecommunication infrastructure (IOT).

The designed biomedical device will be in high demand as the cardiovascular diseases are the major cause of deaths in Pakistan and many medical institutes will be looking to implement this device in their institutes for accurate ECG diagnostic purposes.

Technical Details of Final Deliverable

The final deliverables of this project will be the following sub modules:

1. Sensing System
2. Visualizing Component
3. Central System
4. Cloud Computing
5. Machine Learning Algorithm

1) Sensing System:  Sensing system contains an ECG module that will monitor the person's heart activity.

2) Visualizing Component: This component contains an LCD display attached to central system.

3) Central System: All the above mentioned systems will be integrated with the central system including WIFI sensor, LCD display etc.

4) Cloud Computing: The ECG recordings from the central system will then be sent to the cloud for diagnostic purposes. The cloud will have a machine learning model running in the background to classify ECG signals into normal or abnormal ones. Once classified, the results along with the proposed treatment will be shown on the visualizing component.

Machine Learning Algorithms :

Neural network (NN) finds role in variety of applications due to combined effect of feature extraction and classification availability in deep learning algorithms. Here convolutional Neural Network (CNN) and Recursive Neural Network (RNN) will be utilized to make the device smart enough to classify different ECG signals based on the input signals. 

CNN signal classifications takes an input signal, process it and classify it under certain categories (Eg., Normal, Abnormal). Computers sees an input signal as array of sampling points and it depends on the signal resolution.

The recurrent structure of RNN makes it capable of learning and making full use of the temporal information of the input signals to make up for the deficiencies of the short-term features

Deep learning CNN and RNN models will train and test, each input signal and pass it through a series of convolution layers with filters, Pooling, fully connected layers (FC) and apply different functions to classify the given signals.

---

Overview of the Proposed System

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Health

Other Industries

IT , Medical

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT), Cloud Infrastructure, Others

Sustainable Development Goals

Good Health and Well-Being for People, Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com