Zhiguo Zhou, PhD
Zhiguo Zhou, PhD
Department of Biostatistics & Data Science
University of Kansas Medical Center
University of Kansas Cancer Center (KUCC)
Zhiguo Zhou is an Assistant Professor in Department of Biostatistics & Data Science at University of Kansas Medical Center (KUMC) and Associate Member at University of Kansas Cancer Center (KUCC). He received his B.S. and Ph.D. degrees at Xidian University (Xi’an, China) in 2008 and 2014, respectively. He was a visiting scholar at Leiden University (Leiden, the Netherlands) from May 2013 to May 2014. Since December 2014, he worked as Postdoctoral fellow at Department of Radiation Oncology, UT Southwestern Medical Center (Dallas, TX). Then he was promoted as Research Instructor in September 2017. Before moving to KUMC in May 2022, he worked as an Assistant Professor in School of Computer Science and Mathematics at University of Central Missouri (Warrensburg, MO) starting from August 2019. He has published more than 80 journal or conference papers. He is the editor board member of 2 journals and guest associate editor of Medical Physics. He was the reviewer of more than 20 journals and session chair in multiple conferences. His research interests include radiomics, treatment outcome prediction, medical image processing, reliable artificial intelligence, machine learning and deep learning, knowledge representation and reasoning
Automated Multi-Objective Delta Radiomics for Immunotherapy Response Prediction in Metastatic Melanoma Treatment
Melanoma is the fifth most common cancer in US, with continued rise in incidence. Although there have been many efforts to improve early diagnosis of melanoma, patients with metastatic melanoma (MM) are particular poor. Immunotherapy has demonstrated an increased overall survival and progression-free survival in patients with MM. However, not all the patients have the positive response to the treatment, and some patients even experience serious adverse effects. Even though immune related Response Evaluation Criteria in Solid Tumors (irRECIST) has been developed, it can’t predict pseudo-progression sufficiently. Delta radiomics which contains tumor change that can reflect the progression of treatment is a potential way to predict response in immunotherapy. In this talk, a new developed automated multi-objective delta radiomics (Auto-MODR) model will be presented to overcome the balance issue in current delta radiomics. The real patient studies will be shown to demonstrate model performance as well.