Woodruff School of Mechanical Engineering
Faculty Candidate Seminar
Quantitative CT image formation and deep analysis
Dr. Tianye Niu
Monday, February 25, 2019 at 11:00:00 AM
Boggs Building, Room 3-47
Dr. Steven Biegalski
Precision medicine for diagnosis and treatment relies on the integration of multiple information where imaging plays a fundamental role. As a tomographic imaging modality, x-ray CT and its variant conebeam CT (CBCT) is becoming indispensable in disease diagnosis and treatment guidance. Though widely applied in clinical practice, their imaging performance and interpretation are still evolving. This presentation will mainly describe two significant components of tomographic x-ray techniques: CT image formation and their deep analysis based on clinical requirements. As the foundation of CT imaging, the physics of x-ray photons interacting with matter leads to a suboptimal reconstruction procedure. The model-based reconstruction and artifacts correction will be introduced to achieve clinically acceptable images. To further exploit the physics of x-ray interaction, the material composition of human tissues will be quantified by performing the mathematical optimization on the acquired CT images, also assisted with clinical experience. Deep analysis is applied to the formatted images of CT and other modalities to jointly extract the image features related to the disease evaluation including cancer staging, treatment outcome prediction, complete responder identification, biological correlation, etc. The developed algorithms and protocols will be integrated into the software or hardware tools to facilitate clinical application.
Dr. Niu is an energetic and productive researcher in the systematic approaches to medical physics and engineering. He received his B.S. (2003) and Ph.D. (2009) from the University of Science and Technology of China. He became a postdoctoral research fellow and research scientist under the supervision of Dr. Lei Zhu from 2009-2013 at the Medical Physics Program at Georgia Institute of Technology. Dr. Niu’s career as an independent PI began in 2014 at the Institute of Translational Medicine and Sir Run Run Shaw Hospital, Zhejiang University, Hangzhou, China. Motivated by both academic and clinical application, his research has been focusing on two important aspects of medical physics: a tomographic image formation and disease-related medical image analysis. Specifically, he has investigated various technical tools to achieve precision CT imaging including mathematical reconstruction, image artifacts correction, and material quantification. He has also applied the deep learning technique to the image database (radiomics) of the patient cohort to quantitatively evaluate treatment responses. Starting from 2014, he contributed to 25 peer-reviewed journal publications, 13 of which are the first, senior or corresponding authorship. His total citation is greater than 700 and H-index is 15. He received several research grant awards from national and provincial funding agencies and commercial companies in China due to his high research performance. He is an adjunct associate professor with the Department of Radiation Oncology, Rutgers-The State University of New Jersey. To contribute to the academic community, he also serves as the international advisory board member of Physics in Medicine and Biology and associate editor of the Journal of Applied Clinical Medical Physics. In the past six years at Zhejiang University, Dr. Niu has mentored six undergraduate students, five master students, five Ph.D. students, and two postdocs. After graduation, they have been offered to pursue higher academic achievements in the US (e.g., Johns Hopkins, UT Austin, etc.) or work in large international corporations (e.g., Intel, Huawei, etc.).
Refreshments will be served.