Biomedical AI Extension / Poster Recognition
Radiogenomic MRI Classification
Radiogenomic brain tumor classification from MRI sequences.
This project explores MRI-based radiogenomic classification as a biomedical AI extension of my broader interest in robust visual representation learning, multimodal modeling, and clinically relevant decision support.
Overview
The project studies how MRI information can support radiogenomic prediction while keeping the framing careful: it is presented here as a translational biomedical AI direction and poster-recognized work, not as a peer-reviewed paper.
Problem Statement
Radiogenomic classification seeks to infer clinically relevant molecular or genetic markers from medical imaging. This is challenging because medical datasets can be limited, heterogeneous, and sensitive to acquisition and cohort shift.
Motivation
Reliable biomedical AI requires robust representation learning, careful evaluation, and clinically meaningful model behavior. This direction connects my computer vision background to medical imaging and multimodal decision-support systems.
Methodology
The work is framed around an MRI-based radiogenomic classification workflow. I keep the public description conservative until complete model, evaluation, and material details are ready to share.
Key Contributions
- MRI-based radiogenomic classification workflow.
- Biomedical AI extension of robust visual and multimodal learning.
- Evaluation-oriented framing for clinically relevant prediction.
Experimental Setup
| Input modality | MRI sequences. |
|---|---|
| Task | Radiogenomic brain tumor classification. |
| Status | Biomedical AI extension presented as a poster/presentation item. |
Results / Impact
This work received Top 3 Best Poster recognition at the Pathways 2026 Symposium organized by NextGen Precision Health.
My Role
My contributions include problem framing, multimodal modeling direction, experimental analysis, and poster communication.
Related Presentation
Md Hasibur Rahman. "Radiogenomic Brain Tumor Classification from MRI Sequences." Pathways 2026 Symposium, NextGen Precision Health. Recognized as Top 3 Best Poster.