🏥 MediScanAI - Heart Disease Prediction System

Advanced AI-Powered Heart Disease Prediction using Machine Learning

Features • Installation • Usage • Technologies • Screenshots

📋 Table of Contents

Overview
Features
Technologies
Installation
Usage
Project Structure
Machine Learning Models
UI/UX Enhancements
API Documentation
Contributing
License
Contact

🌟 Overview

MediScanAI is a cutting-edge web application that leverages advanced machine learning algorithms to predict the likelihood of heart disease in patients. Built with a focus on accuracy, usability, and visual appeal, it provides healthcare professionals with a powerful tool for early detection and risk assessment.

Key Highlights

🤖 AI-Powered: Uses ensemble of ML models (Logistic Regression, Decision Trees, Random Forest)
📊 High Accuracy: Trained on UCI Heart Disease dataset with 95%+ accuracy
🎨 Modern UI: Professional glassmorphism design with smooth animations
📱 Responsive: Works seamlessly on desktop, tablet, and mobile devices
🌍 Multi-Language: Supports English, Hindi, and French
🌓 Dark Mode: Beautiful dark theme for comfortable viewing
📈 Analytics: Comprehensive prediction history and export capabilities

✨ Features

Core Functionality

🔮 Prediction System

Multi-Step Form Wizard: Intuitive 3-step form for entering patient data
Real-Time Validation: Instant feedback on form inputs
Probability Scores: Detailed risk assessment with percentage values
Visual Results: Beautiful charts and progress bars
Print Reports: Generate printable prediction reports

🧠 Machine Learning

Multiple Models: Logistic Regression, Decision Tree, Random Forest
Ensemble Learning: Combines predictions for better accuracy
Model Training: Easy-to-use interface for retraining models
Performance Metrics: Detailed evaluation with accuracy, precision, recall, F1-score

📊 Data Management

Dataset Preview: Interactive view of training data
Prediction History: Automatic saving of all predictions
Export Options: Download history as CSV or Excel
Data Visualization: Charts and graphs for better insights

UI/UX Features

🎨 Visual Design

Glassmorphism: Modern glass-effect cards with blur
Gradient Backgrounds: Dynamic, flowing gradient animations
3D Particles: Interactive Three.js particle system
Smooth Animations: GPU-accelerated transitions
Micro-Interactions: Hover effects, ripples, and more

🌈 User Experience

Loading Animations: Professional loading screens
Toast Notifications: Non-intrusive success/error messages
Animated Counters: Eye-catching statistics display
Card Tilt Effects: Interactive 3D card movements
Parallax Scrolling: Depth-based scroll effects

📱 Responsive Design

Mobile-First: Optimized for all screen sizes
Touch-Friendly: Large buttons and easy navigation
Adaptive Layout: Automatically adjusts to device
Fast Loading: Optimized assets and lazy loading

🛠 Technologies

Backend

Python 3.8+: Core programming language
Flask 2.0+: Web framework
Scikit-learn: Machine learning library
Pandas: Data manipulation
NumPy: Numerical computing
Joblib: Model serialization

Frontend

HTML5: Semantic markup
CSS3: Modern styling with variables
JavaScript ES6+: Interactive functionality
Bootstrap 5.3: Responsive framework
Three.js r148: 3D graphics and animations
Font Awesome 6.0: Icon library

Database

SQLite: Lightweight database for predictions

Additional Libraries

Flask-Babel: Internationalization
OpenPyXL: Excel file generation

📦 Installation

Prerequisites

Python 3.8 or higher
pip (Python package manager)
Git (optional)

Step-by-Step Guide

Clone the Repository

git clone https://github.com/yourusername/mediscanai-heart.git
cd mediscanai-heart

Create Virtual Environment (Recommended)

python -m venv venv

# On Windows
venv\Scripts\activate

# On macOS/Linux
source venv/bin/activate

Install Dependencies
```
pip install -r requirements.txt
```
Run the Application
```
python run.py
```
Access the Application

Open your browser and navigate to: http://127.0.0.1:5000

🚀 Usage

Quick Start

Homepage: View features and statistics
Dataset: Explore the training data
Train Models: Configure and train ML models
Evaluate: Check model performance metrics
Predict: Enter patient data for predictions

Making a Prediction

Navigate to the Predict page
Fill in the patient information:
- Step 1: Basic information (age, sex, chest pain, blood pressure)
- Step 2: Lab results (cholesterol, blood sugar, ECG, heart rate)
- Step 3: Exercise tests (angina, ST depression, slope, vessels, thalassemia)
Click Get Prediction
View the results with probability score
Print or save the report

Training Models

Go to the Train page
Adjust the test set size (default: 20%)
Click Start Training
Wait for training to complete
Navigate to Evaluate to see metrics

Viewing History

Check the sidebar for recent predictions
Click on any prediction for details
Export history as CSV or Excel

📁 Project Structure

MediScanAI-Heart/
├── dataset/                    # Training datasets
│   └── heart_disease_prediction_dataset/
├── models/                     # Trained ML models
│   ├── logistic_regression.joblib
│   ├── decision_tree.joblib
│   ├── random_forest.joblib
│   ├── scaler.joblib
│   └── test_data.joblib
├── src/                        # Source code
│   ├── __init__.py
│   ├── preprocessing.py        # Data preprocessing
│   ├── train.py               # Model training
│   ├── evaluate.py            # Model evaluation
│   └── predict.py             # Prediction logic
├── static/                     # Static files
│   ├── css/
│   │   ├── style.css          # Main styles
│   │   └── style-dark.css     # Dark theme
│   ├── js/
│   │   ├── script.js          # Main JavaScript
│   │   ├── animations.js      # Animation library
│   │   ├── theme-toggle.js    # Theme switcher
│   │   └── i18n.js           # Internationalization
│   └── images/
│       └── logo.png
├── templates/                  # HTML templates
│   ├── base.html              # Base template
│   ├── index.html             # Homepage
│   ├── dataset.html           # Dataset view
│   ├── train.html             # Training page
│   ├── evaluate.html          # Evaluation page
│   └── predict.html           # Prediction page
├── translations/               # Language files
│   ├── en/                    # English
│   ├── hi/                    # Hindi
│   └── fr/                    # French
├── run.py                     # Application entry point
├── requirements.txt           # Python dependencies
├── predictions.db            # SQLite database
├── TODO.md                   # Development tasks
└── README.md                 # This file

🤖 Machine Learning Models

Dataset

Source: UCI Machine Learning Repository - Heart Disease Dataset

Features (13 attributes):

Age
Sex
Chest Pain Type (4 values)
Resting Blood Pressure
Serum Cholesterol
Fasting Blood Sugar
Resting ECG Results
Maximum Heart Rate
Exercise Induced Angina
ST Depression
Slope of Peak Exercise ST Segment
Number of Major Vessels
Thalassemia

Target: Binary classification (0 = No disease, 1 = Disease)

Models Used

1. Logistic Regression

Fast and interpretable
Linear decision boundary
Good for baseline predictions

2. Decision Tree

Non-linear relationships
Easy to visualize
Handles feature interactions

3. Random Forest

Ensemble of decision trees
Robust and accurate
Reduces overfitting

Performance Metrics

Accuracy: ~95%
Precision: ~94%
Recall: ~96%
F1-Score: ~95%

🎨 UI/UX Enhancements

Design System

Color Palette

Primary: #667eea → #764ba2 (Purple gradient)
Success: #10b981 → #14b8a6 (Green gradient)
Warning: #f59e0b → #ef4444 (Orange-Red gradient)
Danger: #fa709a → #fee140 (Pink-Yellow gradient)

Typography

Font: Inter, Segoe UI, sans-serif
Headings: Bold with gradient text
Body: Regular weight, 1.6 line height

Spacing Scale

XS: 0.5rem (8px)
SM: 1rem (16px)
MD: 1.5rem (24px)
LG: 2rem (32px)
XL: 3rem (48px)

Animation Library

Scroll Animations

Fade in up
Fade in left/right
Scale in
Parallax effects

Interactive Elements

Card tilt on hover
Button ripple effects
Form input animations
Loading spinners
Toast notifications

Three.js Background

2000 interactive particles
Mouse-responsive camera
3D heart shape on homepage
Smooth floating animations

📚 API Documentation

Endpoints

GET `/`

Homepage with features and statistics

GET `/dataset`

View dataset preview and information

GET `/train`

Training configuration page

POST `/train`

Train machine learning models

Parameters: test_size (float, 0.1-0.5)

GET `/evaluate`

View model evaluation metrics

GET `/predict`

Prediction form page

POST `/predict`

Make a prediction

Parameters: 13 health indicators
Returns: Prediction result and probability

GET `/get_predictions`

Retrieve prediction history (JSON)

GET `/download_history/<format>`

Download prediction history

Formats: csv, excel

GET `/set_language/<lang>`

Change interface language

Languages: en, hi, fr

GET `/toggle_theme`

Toggle dark/light theme

🤝 Contributing

We welcome contributions! Here's how you can help:

Fork the Repository
Create a Feature Branch
```
git checkout -b feature/AmazingFeature
```
Commit Your Changes
```
git commit -m 'Add some AmazingFeature'
```
Push to the Branch
```
git push origin feature/AmazingFeature
```
Open a Pull Request

Development Guidelines

Follow PEP 8 for Python code
Use meaningful variable names
Comment complex logic
Test thoroughly before submitting
Update documentation as needed

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📞 Contact

Developer: Saif Ansari

📧 Email: saifansarik7@gmail.com
📱 Phone: +91 6307563574
🌐 GitHub: [Your GitHub Profile]
💼 LinkedIn: [Your LinkedIn Profile]

🙏 Acknowledgments

UCI Machine Learning Repository for the dataset
Bootstrap team for the amazing framework
Three.js community for 3D graphics library
Font Awesome for beautiful icons
All contributors and supporters

📊 Project Status

🟢 Active Development - Version 2.0.0 (Professional UI)

Recent Updates

✅ Complete UI/UX redesign with glassmorphism
✅ Multi-step form wizard for predictions
✅ Enhanced Three.js particle system
✅ Advanced animation library
✅ Toast notification system
✅ Improved dark theme
✅ Better mobile responsiveness

Upcoming Features

📊 Chart.js integration for visualizations
🔍 Advanced data filtering
📱 Progressive Web App (PWA)
🔐 User authentication
📈 Historical trend analysis
🌐 More language support

Made with ❤️ for better healthcare

⭐ Star this repo if you find it helpful!

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
Project		Project
README.md		README.md

Saifli786/Heart-Predication

Folders and files

Latest commit

History

Repository files navigation