Medical AI & Uncertainty Quantification Researcher
Dhaka, Bangladesh • Target Intake: September 2027 (Master's or PhD in Europe)
Pioneering evidential deep learning, knowledge distillation, and uncertainty-governed triage cascades to transform black-box neural networks into trustworthy, explainable diagnostic systems for clinical medicine.
Research Track, Dean's Honor Roll. Team Lead at DIU GPU Infrastructure Platform, Research Associate at NBTC, and Member at HIRL.
Academic Excellence Award; Executive Committee Member of IT Club & Atomic Science Club.
Advisor: Dr. Md. Ali Hossain (Associate Professor, Dept. of CSE & Director, NBTC)
[1]A. Rubab, M. T. Hasan. “Sample-Efficient Fine-Grained Classification of Cactus Species Using Hybrid Vision Transformers and Knowledge Distillation”. 2025 28th International Conference on Computer and Information Technology (ICCIT), 2025.
PublishedDOI: 10.1109/ICCIT68739.2025.11491549
[2]Abdullah Rubab, Co-Authors. “Explainable Multi-Scale Knowledge Distillation with Uncertainty-Aware Inference for Efficient Brain Tumor MRI Classification”. International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON 2026), 2026.
Accepted[3]Abdullah Rubab, Co-Authors. “An Efficient Attention-Guided Deep Learning Framework with Explainability for Blood Cell Classification”. International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON 2026), 2026.
Accepted[4]Abdullah Rubab, NBTC Research Team. “RAM-ViT: Retrieval-Augmented Evidential Vision Transformer for Cross-Modal Glioma Grading”. Target Q1 Medical Image Analysis Journal / DIU-DoR Poster Presentation 2026, 2026.
Under Review (Q1 Target)[5]Abdullah Rubab, NBTC Research Group. “PUMA-MIL / PUMA-MIL-Lite: Multimodal Multiple Instance Learning for Glioma Detection with Clinical Interpretability”. Target Q1 Medical Informatics / Oncology Journal, 2026.
Under Review (Q1 Target)[6]Abdullah Rubab, Research Collaborators. “MSEA-Net: Multi-Scale Evidential Attention Network with Uncertainty-Governed Triage for Trustworthy Gastrointestinal Endoscopy Classification”. Target Q1 Biomedical Engineering Journal, 2026.
In Progress / Pre-Submission[7]Abdullah Rubab, Clinical Collaborators. “HCC-TACE-Seg: Foundation Model Feature Distillation for Transarterial Chemoembolization Response Prediction”. Targeting Liver Oncology / Medical AI Workshop, 2026.
In ProgressFederated Tri-Modal Diagnostic Platform for Low-Resource Healthcare
A decentralized medical diagnostic platform orchestrating clinical notes, imaging, and tabular vitals under strict differential privacy guarantees.
Multi-Scale Ensemble with Clinician-Facing Explainability
Production-ready PyTorch neuro-oncology pipeline with Grad-CAM visual heatmaps, uncertainty bounds, and automated radiological reporting.
Evidential AI for Proactive GPU Failure Prediction — Hackathon Winner
Real-time telemetry forecasting system for distributed deep learning clusters using evidential neural networks. Won 1st Place at DIU AI Innovation Hackathon 2026.
Campus-Wide GPU Pooling & High-Performance Compute Orchestration
Team Lead & AI/ML Backend Developer for the initiative connecting and pooling DIU's campus GPU resources into an elastic compute cloud.
Multi-Agent System for High-Impact Q1 Academic Workflows
Autonomous multi-agent orchestration framework for literature synthesis, mathematical formalization verification, and experiment tracking.