🧠 Advanced Chest Infection vs Pneumonia Diagnostic Tool

An AI-inspired, multi-modal respiratory diagnostic web application that integrates:

• Symptom-based weighted inference
• Risk factor modeling
• Simulated wearable biometrics
• Lung sound + cough analysis
• Interactive data visualisation
• Voice, audio, and chatbot interfaces

This tool is designed as a research-grade prototype for digital health diagnostics, combining frontend engineering with heuristic clinical reasoning.

🚀 Core Features

🩺 Diagnostic Engine

• Weighted scoring model for:

  • Chest Infection (Bronchitis)
  • Pneumonia

• Dynamic scoring adjustments using:

  • Symptoms
  • Risk factors
  • Simulated wearable data
  • Lung sound analysis

• Output classification:

  • Chest Infection leaning
  • Pneumonia leaning
  • Inconclusive

📊 Data Visualisation (Chart.js)

All charts are:

• Real-time updated
• State-persistent
• Fully resettable

⌚ Wearable Simulation Layer

Simulated biometric inputs:

• Heart Rate
• Body Temperature
• Respiration Rate
• Oxygen Saturation (SpO₂)

Dynamic influence:

• Elevated temperature → increases pneumonia weighting
• Low lung sound score → increases pneumonia likelihood

🎤 Audio & Voice Systems

Cough Recording (Web Audio API)

• 5-second microphone capture
• Simulated cough severity scoring

Voice Command System

• Speech-to-text via Web Speech API
• Automatically selects symptoms based on spoken input

Text-to-Speech Output

• Reads diagnostic result aloud

🤖 Chatbot Assistant

• Lightweight rule-based conversational system

• Context-aware responses:

  • “What” → clinical advice
  • “Why” → explanation of model
  • fallback → evaluation summary

🧾 Reporting & Export

Includes:

• Symptoms
• Risk factors
• Score output
• Timestamp

🧠 Personal Health Journal

• Persistent entries (localStorage)
• Timestamped logs
• Export to TXT

⚙️ Advanced Settings System

• Real-time tuning of:

  • Symptom weights
  • Pneumonia weights
  • Risk factor weights

• Immediate recalculation on save

🌗 UI / UX Features

• Dark mode (persistent)
• Responsive layout
• Animated news ticker
• Modal-based settings panel
• Toggleable health journal

🔔 Smart Notifications

• Periodic browser reminders to re-evaluate symptoms

📡 Sharing Features

• Web Share API integration
• Twitter/X sharing support

🧪 Diagnostic Methodology

The system uses a heuristic weighted inference model:

Total Score = Σ(Symptom Weights) + Σ(Risk Factors) + Dynamic Signals

Dynamic modifiers:

• Temperature > 37°C → Pneumonia++
• Lung score < 5 → Pneumonia++
• Risk stacking → Amplified severity

Trend tracking:

Score Difference = Pneumonia - Chest Infection

Stored over time for longitudinal analysis.

🛠️ Tech Stack

📂 Project Structure

/project-root
│
├── index.html   # Full application (all logic embedded)
├── README.md    # Documentation

⚠️ Disclaimer

This tool is:

• ❌ NOT a medical device
• ❌ NOT clinically validated
• ❌ NOT a substitute for professional diagnosis

It is intended for:

• Educational purposes
• Prototyping
• Research and experimentation

🔬 Future Enhancements

Planned improvements:

• Real ML model integration (TensorFlow.js)
• Real audio classification (cough CNN)
• API-based health data ingestion
• Doctor dashboard + patient profiles
• Cloud sync + authentication
• Real-time anomaly detection

📸 Preview Concept

(Recommended to add screenshots or a demo GIF here in your repo)

📜 License

MIT License – free to use, modify, and distribute.

👨‍💻 Author

Developed by Kai Piper Focused on:

• AI systems
• Simulation environments
• Digital health innovation

⭐ Contribution

Pull requests, forks, and feature suggestions are welcome.

💡 Summary

This project is essentially a hybrid between:

• Clinical decision support system
• Interactive health dashboard
• Experimental AI diagnostic model

It demonstrates how multi-modal data fusion can be implemented purely in the browser.