NeoScreenix: AI-Powered Breast Cancer Detection
🏆 Global Champion at Johns Hopkins Healthcare Design Competition 2025
Project Overview
NeoScreenix is an innovative health-tech solution that combines semi-supervised learning with multimodal data integration for early breast cancer detection. The project has evolved into PinkLifeLine, a startup funded by the Bangladesh National ICT Division.
Key Features
- Semi-Supervised Learning: Leveraging both labeled and unlabeled data for robust classification
- Multimodal Integration: Combining multiple data sources for comprehensive analysis
- Demographic Analysis: Comparative studies across diverse population groups
- Contextualized Vision Pipelines: Advanced computer vision techniques for medical imaging
Achievements
🏆 Global Champion - Johns Hopkins Healthcare Design Competition 2025
🚀 Startup Founded - PinkLifeLine (Co-Founder & Head of Research)
💰 Funding - Bangladesh National ICT Division
🤝 Partnerships:
- Sustainlaunch Labs (Global innovation accelerator)
- Herwill (Women’s empowerment & digital inclusion)
📰 Media Coverage: Featured in Daily Star, TBS News, UNB News
Technical Implementation
- Machine Learning: Deep learning models optimized for medical imaging
- Data Pipeline: Built and leading ML clinical data pipelines & screening workflows
- Medical Network: Coordinating 50+ medical professionals for validation and deployment
- Accessibility: Focused on making advanced screening accessible in resource-limited settings
Research Publications
NSyS 2024: “Breast Cancer Detection Using Semi-Supervised Learning: Multimodal Data Integration and Comparative Demographic Analysis” (Accepted as Poster)
Impact
The project aims to revolutionize early breast cancer detection, particularly in underserved communities, by combining cutting-edge AI with practical healthcare delivery systems.
Supervisor: Prof. Dr. M. Sohel Rahman, BUET
PinkLifeLine is committed to making breast cancer screening accessible, affordable, and accurate for women worldwide.