Advancing Medical Imaging through Explainable and Uncertainty-Aware AI

My research focuses on developing advanced AI and deep learning methods for medical imaging to improve automated disease diagnosis, monitoring, and clinical decision support across modalities such as MRI, CT, ultrasound and ophthalmic imaging.

Craig K. Jones, PhD, is a Professor in the Department of Biomedical Engineering at the University of Alberta, where his work focuses on advancing artificial intelligence and deep learning for medical imaging to improve disease diagnosis, monitoring, and clinical decision support.

His research emphasizes explainable and uncertainty-aware AI, spanning unsupervised and self-supervised representation learning, multi-modal and longitudinal image analysis, radiomics, artifact detection, and federated learning, with applications across MRI, CT, ophthalmic imaging, and image-guided surgery. He has authored more than 100+ peer-reviewed publications, contributed to three patents, and led or co-led projects on topics such as epistemic and aleatoric uncertainty quantification in 3D segmentation, AI-enabled cone-beam CT reconstruction, neuroendoscopic guidance, retinal image analysis, and automated radiomics pipelines for vascular malformations and pancreas cancer risk stratification. He frequently reviews papers for top level journals and is an Associate Editor of the Journal of Medical Imaging. Prior to joining the University of Alberta, he held faculty and research roles in Computer Science, Radiology, Ophthalmology, and Gastroenterology at Johns Hopkins, co-directed a Radiology AI Lab, and directed the Center for the Advancement of Medical AI, a not-for-profit initiative supporting collaborative imaging and large language model applications in healthcare.

Dr. Jones is deeply committed to mentoring, having supervised numerous MSc and PhD trainees and postdoctoral fellows working at the interface of AI and clinical imaging, and he teaches widely on medical imaging, deep learning, and computer vision, including internationally through MICCAI-endorsed AI SPARK Academy programs in Africa and Asia. His work has been recognized with awards such as the CT Meeting Most Innovative Paper Award for combining physics-based models with deep learning and uncertainty, SPIE poster awards, and invited plenary and departmental lectures on medical AI uncertainty, retinal imaging, and AI-driven imaging biomarkers.