🔬 Cancer Diagnosis System

An AI-powered web application for cancer diagnosis using multiple machine learning algorithms, built from scratch in C++ with a modern web interface.

📋 Table of Contents

• Overview
• Features
• Tech Stack
• Architecture
• Getting Started
• Usage
• Project Structure
• API Documentation
• Machine Learning Models
• Screenshots
• Contributing
• License
• Author

🎯 Overview

The Cancer Diagnosis System is a comprehensive, full-stack application that leverages multiple machine learning algorithms to assist in cancer diagnosis based on genetic mutation data. This project demonstrates the implementation of core ML algorithms from scratch in C++, combined with a modern web-based user interface for real-time diagnosis and analysis.

Key Highlights

• 🧠 Four ML Models: Decision Tree, K-Nearest Neighbors, Logistic Regression, and Naive Bayes
• ⚡ High Performance: C++ backend ensures fast processing and low latency
• 🌐 Modern Web UI: Intuitive interface with real-time data visualization
• 📊 Model Comparison: Evaluate and compare multiple ML models side-by-side
• 🔄 Batch Processing: Queue system for processing multiple patients efficiently
• 📈 Comprehensive Metrics: Accuracy, Precision, Recall, and F1-Score evaluation

✨ Features

Core Functionality

• Multi-Model Diagnosis: Run predictions using four different ML algorithms
• Real-time Analysis: Instant diagnosis results with confidence scores
• Data Management: Load and manage patient and genetic data via CSV files
• Batch Queue System: Process multiple patients in a queue for efficient workflow
• Model Evaluation: Comprehensive performance metrics for all models
• Data Visualization: Interactive charts and graphs using Chart.js
• Responsive Design: Modern, user-friendly interface with dark mode support

Advanced Features

• Hash-based Data Mapping: Efficient genetic data lookup using custom hash tables
• Data Preprocessing: Automated data cleaning and normalization
• RESTful API: Clean API design for easy integration
• CORS Support: Cross-origin resource sharing enabled for web access
• Error Handling: Robust error handling and user feedback
• Settings Management: Configurable API endpoints and preferences

🛠️ Tech Stack

Backend

• Language: C++17
• HTTP Server: cpp-httplib v0.28.0
• Build System: CMake 3.10+
• Architecture: RESTful API

Frontend

• HTML5: Semantic markup
• CSS3: Modern styling with animations
• JavaScript: Vanilla JS (no frameworks)
• Chart.js: Data visualization library

Machine Learning

• Algorithms: Implemented from scratch
  • Decision Tree Classifier
  • K-Nearest Neighbors (KNN)
  • Logistic Regression
  • Naive Bayes Classifier

Data Structures

• Hash Tables: Custom implementation for genetic data mapping
• Vectors & Arrays: Efficient data storage and manipulation
• Queues: Patient processing queue system

🏗️ Architecture

┌─────────────────┐
│   Web Browser   │
│   (Frontend)    │
└────────┬────────┘
         │ HTTP/REST API
         │ (JSON)
┌────────▼────────┐
│  C++ HTTP Server │
│   (Port 8080)    │
└────────┬────────┘
         │
    ┌────┴────┐
    │         │
┌───▼───┐ ┌──▼────────┐
│  ML   │ │  Data     │
│ Models│ │ Structures│
└───────┘ └───────────┘

System Flow

1. Data Loading: CSV files are loaded and preprocessed
2. Model Training: ML models are trained on genetic mutation data
3. Diagnosis: Patient genetic data is analyzed by selected model(s)
4. Results: Predictions are returned with confidence scores
5. Visualization: Results are displayed in the web interface

🚀 Getting Started

Prerequisites

• CMake (3.10 or higher)
• C++ Compiler:
  • Windows: Visual Studio 2019+
  • Linux: GCC
  • macOS: Xcode Command Line Tools
• Python 3 (optional, for serving UI)

Installation

1. Clone the repository

   git clone https://github.com/Muhammad-Hashir-Code/Cancer-Diagnosis-System.git
   cd Cancer-Diagnosis-System/CancerDiagnosisSystem

2. Build the project

   # Create build directory
   mkdir build && cd build
   
   # Configure (Windows)
   cmake .. -G "Visual Studio 16 2019"
   
   # Configure (Linux/macOS)
   cmake .. -DCMAKE_BUILD_TYPE=Release
   
   # Build
   cmake --build . --config Release

3. Run the server

   # From project root directory
   ./build/Release/cds_server.exe  # Windows
   ./build/cds_server               # Linux/macOS

4. Open the web interface

   cd ui
   python -m http.server 8000
   # Open http://localhost:8000/index.html in your browser

For detailed setup instructions, see SETUP_GUIDE.md.

📖 Usage

Loading Data

1. Navigate to the "Load Data" section
2. Click "Load Data on Server"
3. Wait for confirmation message
4. Verify data in the Dashboard

Running Diagnosis

1. Go to "Run Diagnosis" section
2. Add patient genetic records (Gene ID and Mutation Score)
3. Select one or more ML models
4. Click "Run Diagnosis"
5. View results with confidence scores

Evaluating Models

1. Navigate to "Evaluate Models" section
2. Click "Evaluate All Models"
3. Compare performance metrics:
   • Accuracy
   • Precision
   • Recall
   • F1-Score

Queue System

1. Add patients to the queue
2. Process queue to run batch diagnosis
3. View results for all queued patients

📁 Project Structure

CancerDiagnosisSystem/
├── data/                      # CSV data files
│   ├── genes.csv             # Genetic mutation data
│   ├── patients.csv          # Patient information
│   └── test_*.csv            # Test datasets
├── headers/                   # C++ header files
│   ├── CancerDiagnosisSystem.h
│   ├── DecisionTreeClassifier.h
│   ├── KNNClassifier.h
│   ├── LogisticRegressionModel.h
│   ├── NaiveBayesClassifier.h
│   └── ...
├── src/                       # C++ source files
│   ├── Server.cpp            # HTTP server implementation
│   ├── CancerDiagnosisSystem.cpp
│   └── [ML Model implementations]
├── ui/                        # Web interface
│   ├── index.html            # Main HTML file
│   ├── script.js             # Frontend logic
│   └── styles.css            # Styling
├── third_party/               # External libraries
│   └── httplib.h             # HTTP server library
├── CMakeLists.txt            # Build configuration
├── SETUP_GUIDE.md            # Detailed setup guide
└── README.md                 # This file

🔌 API Documentation

Endpoints

GET /status

Get system status and data counts.

Response:

{
  "modelsTrained": true,
  "geneticCount": 150,
  "patientCount": 50
}

POST /load

Load data from CSV files.

Request:

{
  "genesFile": "data/genes.csv",
  "patientsFile": "data/patients.csv"
}

Response:

{
  "success": true,
  "message": "Data loaded successfully",
  "geneticCount": 150,
  "patientCount": 50
}

POST /train

Train all ML models.

Response:

{
  "success": true,
  "message": "Models trained successfully"
}

POST /diagnose

Run diagnosis on patient data.

Request:

{
  "model": "decision_tree",
  "geneticRecords": [
    {"geneId": "GENE_001", "mutationScore": 0.85},
    {"geneId": "GENE_002", "mutationScore": 0.72}
  ]
}

Response:

{
  "success": true,
  "model": "decision_tree",
  "prediction": 1,
  "confidence": 0.87,
  "message": "High risk detected"
}

POST /evaluate

Evaluate all models and return metrics.

Response:

{
  "decision_tree": {
    "accuracy": 0.92,
    "precision": 0.89,
    "recall": 0.91,
    "f1Score": 0.90
  },
  "knn": { ... },
  "logistic_regression": { ... },
  "naive_bayes": { ... }
}

🤖 Machine Learning Models

Decision Tree Classifier

• Type: Supervised Learning
• Use Case: Classification based on feature thresholds
• Advantages: Interpretable, handles non-linear relationships
• Implementation: Custom C++ implementation with recursive splitting

K-Nearest Neighbors (KNN)

• Type: Instance-based Learning
• Use Case: Classification based on similarity
• Advantages: Simple, effective for non-linear data
• Implementation: Distance-based classification with configurable k

Logistic Regression

• Type: Statistical Learning
• Use Case: Binary classification with probability estimates
• Advantages: Fast, interpretable coefficients
• Implementation: Gradient descent optimization

Naive Bayes

• Type: Probabilistic Classifier
• Use Case: Classification based on Bayes' theorem
• Advantages: Fast, works well with small datasets
• Implementation: Gaussian Naive Bayes for continuous features

📊 Data Format

genes.csv

Gene_ID,Mutation_Score,Label
GENE_001,0.85,1
GENE_002,0.72,1
GENE_003,0.45,0

patients.csv

Patient_ID,Name,Age
P001,John Doe,45
P002,Jane Smith,52

🎨 Screenshots

Note: Add screenshots of your application here

• Dashboard showing system statistics
• Diagnosis interface with model selection
• Results visualization with charts
• Model evaluation comparison

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

Contribution Guidelines

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📝 License

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

👤 Author

Muhammad Hashir

• GitHub: @Muhammad-Hashir-Code
• Email: siddiquimuhammadhashir1@gmail.com

🙏 Acknowledgments

• cpp-httplib - HTTP server library
• Chart.js - Data visualization library
• CMake community for excellent build system documentation

📚 Additional Resources

• Setup Guide - Detailed installation instructions
• Data Loading Troubleshooting - Common issues and solutions
• CMake Documentation

⭐ Show Your Support

If you find this project helpful, please consider giving it a ⭐ on GitHub!

---

Built with ❤️ using C++ and modern web technologies