Ph.D. student in Computer Science at Old Dominion University. Building trustworthy, interpretable, and privacy-preserving machine learning systems for healthcare and beyond.
Investigating deep neural network vulnerabilities to adversarial perturbations. Designing robust training procedures and certified defenses for safety-critical deployments in medical imaging.
RobustnessPrivacy-preserving collaborative learning across distributed clients. Addressing data heterogeneity, communication efficiency, and Byzantine-resilient aggregation in healthcare settings.
PrivacyIntegrating interpretability into deep learning pipelines — from vision transformers to ensemble networks — to enable transparent, decipherable predictions in clinical decision support.
InterpretabilityApplying deep learning to blood cell classification, brain tumor detection, lung disease diagnosis, and retinal imaging for neurodegenerative disease screening.
HealthcareOld Dominion University, USA · GPA: 4.0/4.0
Brac University, Bangladesh · GPA: 4.0/4.0
Brac University, Bangladesh · GPA: 3.74/4.0
Department of Computer Science, Old Dominion University
Department of CSE, University of Liberal Arts Bangladesh (ULAB)
Department of CSE, Brac University
Chungbuk National University, South Korea · Glass-Free 3D Internet TV project
Computer Vision and Intelligent Systems Research Lab, Brac University
Earthquake resistance evaluation using image processing & ML. PI: Md. Ashraful Alam, PhD
iCACCESS 2024 — Int. Conference on Advances in Computing, Communication, Electrical & Smart Systems
Brac University — Highest CGPA in M.Sc. program
ULAB — For Scopus-indexed publications
I'm always open to discussing research ideas, potential collaborations, or opportunities in adversarial ML, federated learning, and medical image analysis. Feel free to reach out.