Title: From Pixels to Prognosis: Building Interpretable AI Biomarkers for Personalized Clinical Risk Prediction
Speaker Bio
Taofik Ahmed Suleiman is a joint PhD student in Biomedical Engineering at the Georgia Institute of Technology and Emory University, where he conducts research at the intersection of artificial intelligence, medical imaging, and precision oncology. His work focuses on developing interpretable AI biomarkers from routine CT imaging to characterise systemic host health, including organ texture, body composition, and biological ageing, and leveraging these biomarkers to predict disease risk and clinical outcomes beyond tumor-centric models.
Prior to his PhD, he earned a master’s degree in Medical Imaging and Applications in Europe and worked as a visiting research scholar at Duke University. His research interests include multi-organ radiomics phenotyping, imaging-derived biological age, survival analysis, and clinically interpretable AI systems, with a particular emphasis on building robust, generalisable models that can be translated across diverse populations and healthcare settings.
Abstract:
Recent advances in artificial intelligence have enabled the extraction of rich, quantitative biomarkers from routine medical imaging, creating new opportunities for personalised clinical risk prediction. This talk provides an overview of how interpretable AI imaging biomarkers can be built across the full pipeline from raw pixels to prognostic models while moving beyond traditional tumor-centric approaches toward imaging the individual. Key concepts in feature extraction, organ and tissue segmentation, and outcome prediction will be reviewed to highlight how imaging-derived phenotypes contribute to disease risk and clinical outcomes. Emphasis will be placed on model interpretability, validation, and common pitfalls that limit clinical translation. The session will also include a brief demonstration to equip participants with practical principles for developing robust, explainable imaging AI models that support personalized clinical decision-making.
Schedule:
Date: Friday, January 16th 2026
Time: 07:00 PST | 16:00 West Africa Time | 15:00 UTC
Zoom Meeting Link: https://us06web.zoom.us/j/89709172931
Contact Information
compass@mailab.io
View recordings of previous sessions on our YouTube channel https://www.youtube.com/@COMPASS-nn5or
Leave a Reply