Date/Time
Date(s) - 03/31/2025
3:00 pm - 4:00 pm

Location
Communicore, C1-004

Abhinav K. Jha, PhD., Associate Professor of Biomedical Engineering and of Radiology, Washington University

Quantitative imaging is showing strong promise in multiple pre-clinical and clinical applications. For clinical translation of this promise, techniques to clinically evaluate QI methods based on how reliably they measure the underlying true quantitative value are needed. However, clinical evaluation requires availability of a gold standard, which is often unavailable. In this talk, we will discuss the emerging no-gold-standard evaluation techniques that are being developed to address this challenge. These techniques, that are grounded in mathematically rigorous formalisms, have demonstrated the ability to evaluate quantitative imaging methods even in the absence of true quantitative values across multiple imaging modalities including diffusion MRI, PET, and SPECT. We will demonstrate the application of these techniques to assess algorithms we have developed in our lab, including deep-learning methods for image segmentation and quantitative SPECT methods for alpha-particle radiopharmaceutical therapies. These case studies will illustrate how no-gold-standard evaluation techniques could help us realize the goal of evaluating quantitative imaging methods without ground truth.

Biography: Dr. Abhinav Kumar Jha is an Associate Professor at Washington University in St. Louis with joint appointments in Biomedical Engineering and the Mallinckrodt Institute of Radiology and courtesy appointment in the Department of Electrical & Systems Engineering and Department of Computer Science and Engineering. Previously, he was a faculty at Johns Hopkins University. Dr. Jha obtained his PhD with valedictorian honors from the University of Arizona. He is a recipient of multiple awards including the NSF CAREER, NIBIB Trailblazer, Distinguished Investigator Award of the Academy for Radiology and Biomedical Imaging, Young Investigator Symposium Award of Distinction for Translational Sciences at the ECOG-ACRIN meeting and Michael B. Merickel Best Student Paper award at SPIE Medical Imaging. He is the current Chair of the Society of Nuclear Medicine and Molecular Imaging (SNMMI) AI task force and previously led the evaluation team within this task force that proposed the RELAINCE guidelines to evaluate AI algorithms for nuclear medicine. He is on the Board of Directors of the Physics, Instrumentation and Data Sciences Council of the SNMMI, the Board of Directors of the Indo-American Society of Nuclear Medicine and a senior member of the IEEE and SPIE.