Sakib Mostafa, PhD
Postdoctoral Fellow

Stanford Email 
CAP Profile

Sakib is a postdoctoral research fellow at Stanford University with expertise in machine learning, deep learning, and integrative omics analysis. His research focuses on developing AI-driven tools for the integration of multi-omics data, with applications in cancer diagnosis and precision medicine. Prior to joining Stanford, Sakib was a Postdoctoral Associate at the National Research Council of Canada, where he contributed to large-scale genomic projects involving plant pangenomes and trait mapping. He earned his Master’s and Ph.D. from the University of Saskatchewan, Canada.

Seraphina Shi, PhD
Postdoctoral Fellow

Stanford Email 
CAP Profile

Seraphina is a postdoctoral scholar in the Department of Radiation Oncology at Stanford University School of Medicine, co-mentored by Dr. Md Tauhidul Islam and Dr. Mohammad Esfahani. She received her PhD in Biostatistics from UC Berkeley, where she focused on statistical machine learning and computational oncology. At Stanford, her research centers on developing robust and interpretable statistical machine learning methods for complex biomedical data to drive patient-centered impacts, with applications in early cancer diagnostics through cfDNA-based liquid biopsy, spatial transcriptomics, and imaging-based biomarker discovery.

Ridvan Yesiloglu 
Ph.D. Student

Ridvan Yesiloglu is a Ph.D. student in the Electrical Engineering Department at Stanford University, advised by Prof. Ehsan Adeli and Prof. Md Tauhidul Islam. His research focuses on advancing our understanding of the human brain through artificial intelligence. He aims to develop large-scale, clinically transformative brain models that are scalable and widely adoptable in real-world healthcare settings. Ridvan received his B.S. from Bilkent University and is a recipient of the Fulbright Scholarship.

Md Atik Ahamed 
Postdoctoral Fellow

Atik is a postdoctoral scholar at Stanford University with expertise in deep learning, multimodal learning, and real-world data challenges. He earned his Ph.D. in Computer Science from the University of Kentucky and his B.Sc. from Rajshahi University of Engineering & Technology. His research focuses on building highly effective AI solutions and foundational models for healthcare, achieving broad impact across both industry and academia. His work has led to publications in venues such as ICML, AAAI, KDD, ECAI, Nature Communications, Information Fusion, Journal of Biomedical Informatics, and Medical Image Analysis. During his PhD, he interned at Google, leading development of foundational models and benchmarks. He was also selected for the exclusive Meta PhD Forum.

Ucchwas Talukder
Postdoctoral Fellow

Ucchwas is a postdoctoral scholar in the Department of Radiation Oncology at Stanford University School of Medicine. His research focuses on artificial intelligence and biomedical signal processing, with applications in cardiac monitoring, disease detection, respiratory analysis, and silent speech recognition. His current interests include multimodal biomedical reasoning and the development of interpretable representations that connect physiological data, biological networks, and clinical knowledge. His goal is to build accurate, transparent, and dependable AI systems that support scientific discovery, assistive technologies, and clinical research. Ucchwas earned his Ph.D. in Computer Science from Texas Tech University and his B.Sc. from the Bangladesh University of Engineering and Technology.

Tracy Xue 
Assistant Clinical Research Coordinator

Tracy is an Assistant Clinical Research Coordinator in the Islam Lab. She received her B.S./M.S. in Bioengineering from UCLA in 2025, with her background spanning neurodegenerative disease research at Washington University in St. Louis, nephrotoxicity modeling at Genentech, and contractility-based drug screening platform development at Forcyte Biotechnologies. In the Islam Lab, she works on applying Graph Neural Networks that integrate biological pathway structure with multi-omics cancer data to better understand the complex interactions driving tumor behavior.