Dr. Julius Bamwenda is a researcher and engineer specializing in artificial intelligence for biomedical image analysis, with expertise in computational pathology, digital hematology, and deep learning for medical imaging. His research focuses on developing advanced AI frameworks for the detection, segmentation, and analysis of cells in microscopic and whole-slide images using state-of-the-art deep learning and foundation models. He is the lead author of the recently published SCI-indexed paper, "TransNet-SAM2: A Transformer–Foundation Model Framework for Prompt-Free Segmentation of White Blood Cells in Microscopic Blood Smear Images," which presents a new framework that combines transformer-based feature learning with the Segment Anything Model 2 (SAM2) for accurate, prompt-free white blood cell segmentation. His work also includes the development of hybrid deep learning architectures that combine object detection, instance segmentation, and foundation models to improve the robustness, accuracy, and clinical applicability of automated hematological image analysis. Dr. Bamwenda's research spans computational pathology, explainable artificial intelligence, weakly supervised learning, diffusion models, vision transformers, and multimodal foundation models for medical image analysis. His long-term goal is to develop trustworthy and clinically deployable AI systems that enhance diagnostic accuracy, reduce annotation effort, and support precision medicine in hematology and digital.
Dicle University
Diyarbakır, Turkey (Türkiye)
Electrical and Electronics Engineering