BookMyDoc - Doctor Appointment System

Introduction

BookMyDoc is a web-based doctor appointment system designed to streamline medical consultation scheduling. It allows patients to book appointments with doctors, receive updates, and leverage machine learning for disease prediction based on symptoms.

• Live Demo: BookMyDoc - Live [Backend deployed on Render free tier, may take a moment to load initially]

---

Key Features

• User Authentication: Secure login and signup using JWT with cookie expiration.
• Doctor & Patient Selection: Users can register as either a doctor or a patient.
• Appointment Management:
  • Doctors can view patient details (name, date, email, status: pending, confirmed, completed, or canceled).
  • Patients can request appointments and receive notifications.
  • Calendar view with different status colors for easy tracking.
• Email Notifications: Patients receive an email when a doctor confirms or cancels an appointment. The email includes the appointment status and a QR code for verification.
• Verification Appointment:
  • Patients receive an email with the approval or cancellation status and a QR code.
  • When both the doctor and patient meet, the doctor scans the QR code to verify the appointment.
• Doctor Review:
  • Patients receive an email link to rate their experience with a star rating and comments.
  • Reviews help future patients make informed decisions and assist doctors in improving service quality.
• Disease Prediction:
  • Three machine learning models implemented:
    • General Disease Prediction:
      • Uses symptoms to predict diseases.
      • Trained on 377 symptoms and 200 diseases.
      • Implements Logistic Regression with an 86% accuracy score.
    • Diabetes Prediction For Women:
      • Uses patient details such as BMI, insulin, glucose levels, and pregnancy history.
      • Determines if a patient is diabetic (positive or negative result).
      • Implements AdaBoost with a 79% accuracy score.
    • Stroke Prediction:
      • Uses features like gender, age, work type, and various health parameters.
      • Predicts whether a user is at risk of stroke.
      • Implements Gradient Boosting with a 91% accuracy score.

---

• Website Demo Video (Youtube link):
  
  

---

Technologies Used

Frontend

• HTML, CSS, Tailwind CSS - Styling and UI components.
• React.js - Dynamic frontend framework for an interactive UI.
• Deployed on Vercel.

Backend

• Node.js & Express.js - Server-side development.
• MongoDB (Atlas) - NoSQL database for storing user and appointment data.
• JWT - Secure authentication and session management.
• Deployed on Render.

Machine Learning

• Python & Flask - API integration for disease prediction.
• Scikit-learn - Used for training and implementing ML models.
• Models Implemented:
  • Logistic Regression: General disease prediction.
  • AdaBoost: Diabetes prediction for women.
  • Gradient Boosting: Stroke prediction.
• Deployed on Render.

---

How It Works

1. User Registration & Login

   • Sign up as a doctor or patient.
   • Authentication is handled using JWT for security.

2. Appointment Booking & Management

   • Patients browse and request appointments with doctors.
   • Doctors can accept or decline appointment requests.
   • Status updates (Pending, Confirmed, Completed, Canceled) are reflected in real time.
   • Calendar view displays appointment statuses for easy tracking.

3. Verification Appointment

   • Patients receive an email with the approval or cancellation status and a QR code.
   • When both the doctor and patient meet, the doctor scans the QR code to verify the appointment.

4. Disease Prediction

   • The main feature is disease prediction using symptoms.
   • Additionally, we have separate models for Diabetes Prediction (for women) and Stroke Prediction.

---

Installation & Setup (run locally)

1. Clone the repository:

   git clone https://github.com/ParamPS25/DoctorAppointmentSystem.git
   cd DoctorAppointmentSystem

2. Set up Environment variables
   

   Frontend (.env)

   Create a .env file in the frontend directory with the following variables:

   VITE_ML_DOC_API = http://localhost:5000
   VITE_NODE_DOC_API = http://localhost:8080

   Backend (.env)

   Create a .env file in the backend directory with the following variables:

   MONGO_URI = mongodb+srv://username:password@your-cluster.mongodb.net/your-database
   PORT = 8080
   JWT_ACCESS_SECRET = your_jwt_access_secret_key
   JWT_REFRESH_SECRET = your_jwt_refresh_secret_key
   JWT_EXPIRES_IN = 1h
   COOKIE_EXPIRES_IN = 24
   MAIL_PASS = your_email_app_password
   FRONTEND_URL=http://localhost:5173

3. Install dependencies for the frontend:

   cd frontend
   npm install
   npm run dev

4. Install dependencies for the backend:

   cd backend
   npm install
   npm run dev

5. Run the machine learning backend:

   cd ml-backend
   pip install -r requirements.txt
   python app.py

6. Using Conda for ML Backend (Alternative Setup):

    cd ml-backend
    conda create -n doc_env python=3.8 -y
    conda activate doc_env
    pip install -r requirements.txt
    python app.py

---

Contributors

• Param Bhavasar - Backend Development
• Purv Patel - Frontend Development
• Prit Patel - Frontend Development
• Prince Patel - Machine Learning

---

License

• This project is licensed under the MIT License.