Virtual Consultant

Project Title

Virtual Consultant

Project Area of Specialization Artificial IntelligenceProject Summary

Virtual Consultant is an AI-based Web App that will assist medical experts to predict if the patient has Covid-19 disease or any heart disease with the help of trained Machine Learning and Deep Learning Models. Detecting Heart Disease (cardiovascular disease) and Covid-19 is a very complex task involving a lot of factors. The heart Disease Model takes 10 features including Gender, Age, Cholesterol level, Blood Pressure as input for predicting Heart Disease. While Covid19 Model takes 8 features including Cough, Fever, Shortness of breath, Headache, and an XRay image as input for predicting the Covid19.

Different Algorithms like Random Forest, Logistic Regression, Decision Tree, Fully Connected Dense layers, and Multi-Input Modal were tested to obtain high accuracy with the training and testing data. 

Project Objectives

• To detect the unnoticed diseases at the initial stages.
• To develop and train the deep learning models on heart disease and covid 19 datasets.
• To evaluate the performance of the proposed algorithms/ Models.
• To make an easily accessible and usable interface for users

Project Implementation Method

• Data Collection and Preprocessing
  • The first step is to collect the data. The data for the Covid-19 detection including 5000 - 6000 X-ray images and tabular data was collected from Kaggle and GitHub respectively and Heart Disease data was collected from GitHub. The next step is to preprocess the data (Handling Null Values, Outlier Detection and Removal, Normalizing the X-ray images and tabular data) to make it feasible for Machine Learning Models.
• Model Training and Testing
  • Heart Disease Prediction
    • Machine Learning
      • We used different Machine Learning  Algorithms such as Random Forest, Logistic Regression, Multi-Layer Perceptron to train and test the model.
    • Deep Learning Model
      • We used a simple Model with Fully Connected Dense Layers and Batch Normalization Layers. RMSprop optimizer, Binary Cross entropy loss function, 64 Batch Size, 1000 Epochs, ReLU, and Sigmoid activation functions were used.
  • Covid 19 Detection
    • Deep Learning Model
      • We used Multi-Input Modal that takes features and XRay image as an input to predict the Covid 19. RMSprop optimizer, Binary Cross entropy loss function, 64 Batch Size, 10 Epochs, ReLU, and Sigmoid activation functions were used.
• Deployment
  • We used HTML, CSS, Javascript for the Front End, and Python for the Back End to deploy the Web Application.

Benefits of the Project

The Virtual Consultant Web App will help experts for easy and AI-based detection of the disease at early stages. These diseases are usually hard to find in the early stages. 

By early detection of Heart Disease and Covid-19, we can reduce the risk of Mortality greatly by taking cautious measures after detection.

This Web App will also help normal people (people that do not have knowledge of medical science), and people living in remote areas to diagnose and detect Heart Disease and Covid19 by themselves where the doctors are not easily available.

We'll add a very easy-to-use user interface including text box inputs for symptoms like Cough, Fever, cholesterol, etc. to the Virtual Consultant Web App.

Technical Details of Final Deliverable

The final deliverable of the project will be a Web App named Virtual Consultant. The front end of the Virtual Consultant is based on front-end technologies HTML, CSS, JavaScript, and a CSS and JS Framework, Bootstrap. The Backend of the Virtual Consultant is based on a micro web framework, Flask, written in Python. We will use a custom API for integrating the Models with Web App written in Flask Python.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries Medical , Health Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Nvidia GPU - 4GB	Equipment	1	70000	70000
Web Hosting & Internet Services	Miscellaneous	1	10000	10000