A smart, AI-powered mobile application for early breast cancer risk prediction using TensorFlow Lite, OCR, and image-based analysis.
- AI-powered prediction using tabular ML model
- Image-based prediction using CNN (TensorFlow Lite)
- OCR-based report analysis (Google ML Kit)
- Automatic extraction of values from medical reports
- Form-based prediction with clinical parameters
- Quick Self-Check for awareness and early detection
- Displays prediction with confidence score
- Works offline with on-device ML
- Clean and intuitive Material UI
- Secure user authentication
- Educational section for breast cancer awareness
- Error handling and input validation
- Frontend: Android (Java, XML)
- Machine Learning: TensorFlow Lite
- OCR: Google ML Kit
- Architecture: Modular Android Architecture
- Build: Gradle
- IDE: Android Studio
- User opens the application
- Chooses prediction method:
- Manual data entry
- Upload medical report (OCR)
- Upload image
- Quick Self-Check
- Data is preprocessed and normalized
- TensorFlow Lite model performs prediction
- Result displayed with classification and confidence score
app/
├── java/com/breastscan/
├── assets/ (TFLite models)
├── res/ (UI layouts, drawables)
├── ml/ (model handling classes)- Android Studio (latest)
- Android SDK installed
- Java 8+
git clone https://github.com/Smriti-Prajapati/BreastScan.git
cd BreastScan- Open project in Android Studio
- Sync Gradle
- Connect Android device / Emulator
- Run the app
- User Authentication Module
- Form-Based Prediction Module
- OCR Report Analysis Module
- Image-Based Prediction Module
- Quick Self-Check Module
- Result Display Module
- User Profile & Menu Module
- Fast and real-time predictions
- Supports multiple input methods
- Reduces manual effort using OCR
- Works offline (no internet required)
- Promotes awareness and early detection
- Easy-to-use interface
- Accuracy depends on input data and model
- OCR may fail on low-quality images
- Not a replacement for professional medical diagnosis
- Improve model accuracy with larger datasets
- Cloud integration for advanced analytics
- iOS version support
- Telemedicine integration
- Multilingual support
Smriti Prajapati
This application is intended for educational and preliminary assessment purposes only. It should not be used as a substitute for professional medical advice or diagnosis.