Prof. Dinggang Shen

School of BME, ShanghaiTech University, China

Shanghai United Imaging Intelligence Co., Ltd.

AI based Medical Image Reconstruction

Bio:Dinggang Shen is Professor and Founding Dean of School of Biomedical Engineering, ShanghaiTech University, and also Co-CEO of United Imaging Intelligence (UII). He is Fellow of IEEE, Fellow of The American Institute for Medical and Biological Engineering (AIMBE), Fellow of The International Association for Pattern Recognition (IAPR), and also Fellow of The Medical Image Computing and Computer Assisted Intervention (MICCAI) Society. He was Jeffrey Houpt Distinguished Investigator, and (Tenured) Full Professor in the University of North Carolina at Chapel Hill (UNC-CH), directing The Center of Image Analysis and Informatics, The Image Display, Enhancement, and Analysis (IDEA) Lab, and The Medical Image Analysis Core. He was also a tenure-track Assistant Professor in the University of Pennsylvanian (UPenn) and a faculty member in the Johns Hopkins University. His research interests include medical image analysis, computer vision, and pattern recognition. He has published more than 1470 peer-reviewed papers in the international journals and conference proceedings, with H-index 121 and >60K citations. He serves as Editor-in-Chief for Frontiers in Radiology, as well as editorial board member for eight international journals. Also, he has served in the Board of Directors, The Medical Image Computing and Computer Assisted Intervention (MICCAI) Society, in 2012-2015, and was General Chair for MICCAI 2019.

Abstract:This talk will introduce various deep learning methods we developed for fast MR acquisition, low-dose CT reconstruction, and low-cost and low-dose PET acquisition. The implementation of these techniques in scanners for real clinical applications will be demonstrated. Also, comparisons with state-of-the-art acquisition methods will be discussed.

Prof. Al Bovik

Director, Laboratory for Image and Video Engineering

University of Texas at Austin, USA

Picture Quality Prediction Outside the Visible Spectrum

Bio:Al Bovik is the Cockrell Family Regents Endowed Chair Professor at The University of Texas at Austin. His research interests land at the nexus of visual neuroscience and digital pictures and videos, particularly regarding how human viewers respond to visual content. An elected member of the U.S. National Academy of Engineers, his international honors include the 2022 IEEE Edison Medal, the 2019 Progress Medal of the Royal Photographic Society, the 2019 IEEE Fourier Award, the 2017 OSA Edwin H. Land Medal, a 2015 Primetime Emmy Award and a 2021 Technology and Engineering Emmy Award from the Academies of Television Arts and Sciences, and the Norbert Wiener Award of the IEEE Signal Processing Society.

Abstract: In this talk I will discuss recent research on the development of no-reference or blind picture quality prediction models, for pictures captured outside of the visible spectrum. Very good algorithms already exist that accurately predict the perceptual quality of still optical pictures, most of them predicated on distortion-induced violations of natural scene statistics (NSS) models, deep learning models, or combinations of these. Here I will discuss the quality prediction of non-optical images, but in the context of task prediction on x-ray screening images. It is my hope this work will stimulate promising avenues of future research.

Dr. Matt Turek

Program Manager, Information Innovation Office (I2O)


Media Authentication and Explainable AI: Applications Beyond the Visible Spectrum

Bio:Dr. Matt Turek joined DARPA's Information Innovation Office (I2O) as a program manager in July 2018, and served as Acting Deputy Director of I2O from June 2021 to October 2021. Dr. Turek is the program manager for DARPA’s Explainable AI (XAI), Machine Common Sense, Semantic Forensics (SemaFor), Media Forensics (MediFor), and In the Moment (ITM) programs. Prior to his position at DARPA, Turek was at Kitware, Inc., where he led a team developing computer vision technologies. His research focused on multiple areas, including large scale behavior recognition and modeling; object detection and tracking; activity recognition; normalcy modeling and anomaly detection; and image indexing and retrieval. His research interests include computer vision, machine learning, artificial intelligence, and their application to problems with significant societal impact.

Abstract:Dr. Turek’s program portfolio at DARPA includes research programs in foundational AI, such as the Machine Common Sense (MCS) and Explainable AI (XAI) programs, and programs in media authentication, such as the Media Forensics (MediFor) and Semantic Forensics (SemaFor) programs. The goal of the XAI program was to develop machine learning techniques that were more understandable and explanation interfaces that could support the explanation process with a user. MediFor and SemaFor seek to develop capabilities to detect, attribute, and characterize falsified media. While the goal of these programs is to build foundational capabilities, there have been applications to domains of interest to PBVS, including commercial satellite data, biological assays, and medical imaging. This talk will provide an overview of the XAI, MediFor, and SemaFor programs and highlight applications to domains of interest to PBVS.