🌬️ CoughLens AI — Voice-First Respiratory Screening Platform

Multilingual • AI-Powered • Cough-Based Diagnosis • Built for Rural India

---

📖 Overview

CoughLens AI is a multilingual, voice-driven diagnostic assistant that analyzes cough audio to detect:

• Asthma
• Pneumonia
• Healthy

Designed for rural healthcare, ASHA workers, and clinics, enabling:

• Full voice navigation
• ML-powered cough detection
• PDF report generation
• IPFS storage
• WhatsApp & Email sharing

Users simply say:

“Start a cough test.”

---

✨ Key Features

🎤 1. Voice Command Center

Hands-free interface supporting Telugu, Hindi, English.

You can:

• Start diagnostic tests
• Add patient information
• Access previous reports
• Ask for analytics
• Navigate dashboards
• Send reports via WhatsApp

---

🤖 2. AI-Powered Cough Analysis

Audio ML pipeline includes:

• Noise reduction
• Segmentation
• MFCC extraction
• Mel Spectrogram generation
• CNN / YAMNet inference

Outputs:

• Disease prediction
• Confidence score
• Acoustic biomarkers
• Voice explanation

---

🗂 3. IPFS-Based Medical Reports

All medical PDFs are:

• Tamper-proof
• Permanent
• Shareable
• Stored using IPFS CIDs

---

🏥 4. Role-Based Dashboards

User	Features
Individual User	Run tests, view history
ASHA Worker	Add patients, run tests via voice
Clinician / Hospital	Case trends, analytics, patient reports

---

🧠 Tech Stack

Frontend

• React.js
• Vite
• Tailwind CSS
• Web Audio API

Backend

• FastAPI
• MongoDB Atlas
• JWT Authentication

AI / Machine Learning

• Python
• librosa
• MFCC + Mel Spectrogram
• CNN / YAMNet

Voice AI

• OpenAI Realtime Voice API
• Whisper STT
• Multilingual TTS

Storage

• IPFS (Web3.Storage / Pinata)

---

👨‍💻 Team

Name	Role
Marka Sai Charan	Full Stack Developer / AI Integration
Maitry Patel	Machine Learning and Cloud
Tanmaya Bhanja	Frontend and Project Management, App Dev
Nidhi Thakore	DevOps and Cloud

---

🔮 Future Enhancements

Phone-call based diagnosis (IVR)

Offline-first mobile app

Multi-disease respiratory classifier

---

🛡️ License

MIT License © 2025 CoughLens AI Team

---

🏗️ Architecture (Mermaid Diagram)

flowchart TD
    A["User<br/>(Voice / Web)"] --> B["Frontend UI<br/>React + Vite<br/>Voice Recorder"]
    B --> C["OpenAI Realtime API<br/>STT • TTS • Intent Engine"]
    C --> D["Backend<br/>FastAPI<br/>Auth • ML Orchestration • Reports"]
    D --> E["MongoDB Atlas<br/>Users • Patients • Reports"]
    D --> F["ML Engine<br/>MFCC • MelSpectrogram • CNN / YAMNet"]
    F --> G["IPFS Storage<br/>Decentralized PDF Reports"]

Loading