Medical Consultation Platform : Ultra Doc

This is a Flask-based web application that allows doctors and patients to connect for medical consultations, book appointments, and conduct video meetings.

Features

• Doctor registration

• Doctor rating system

• Patient appointment booking

• Video consultation via ZegoCloud API

• Medical image classification for disease detection

• Email notifications for appointment confirmations

Technologies Used

• Backend: Flask, PyMongo, TensorFlow, JWT, SMTP (Email)

• Frontend: HTML, CSS, JavaScript

• Database: MongoDB

• Cloud Services: ZegoCloud for video conferencing

Installation

Prerequisites

• Python 3.x

• MongoDB (local or Atlas)

• Flask and dependencies

• TensorFlow for ML model inference

Steps

1. Clone the repository:

   git clone https://github.com/your-repo/medical-app.git
   cd medical-app
   

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate   # On Windows use `venv\Scripts\activate`
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Set up environment variables (or update config in code):

   • MongoDB URI

   • ZegoCloud Credentials (APP_ID & SERVER_SECRET)

   • Email credentials (Gmail SMTP)

5. Run the application:

   python app.py
   

API Endpoints

Endpoint Method Description / GET Home page /doctor_register GET/POST Register a doctor /doctor_consultation GET View list of doctors /book_appointment POST Book an appointment /ml_detection GET Access ML detection page /upload POST Upload medical images for classification /video_meet/<room_id> GET Join a video consultation /generate-zego-token GET Generate authentication token for ZegoCloud

Machine Learning Models

The application uses three pre-trained models:

1. Organ Classification (Mobilenet)

2. Breast Cancer Detection (EfficientNetB7)

3. Ovarian Cancer Detection (PCOS)

Models are loaded from local directories. Ensure the correct paths are provided in the code.

Notes

• Ensure proper security measures for database and email credentials.

• MongoDB indexes should be optimized for query performance.

• ML model paths should be updated according to deployment needs.

License

This project is open-source and available under the MIT License.