Prof. Yudong Zhang
Fellow of IET, Fellow of EAI, and Fellow of BCS, The University of Leicester, UK
Prof. Yudong Zhang worked as a postdoc from 2010 to 2012 at Columbia University, USA, and as an Assistant Research Scientist from 2012 to 2013 with the Research Foundation of Mental Hygiene (RFMH), USA. He served as a Full Professor from 2013 to 2017 at Nanjing Normal University. Now he serves as a Chair Professor at the School of Computing and Mathematical Sciences, University of Leicester, UK. He is also (Honorary) Professor at universities in China, India, and Saudi Arabia. His research interests include deep learning and medical image analysis.
He is the Honorary Follow of World Leadership Academy, Fellow of IET, Fellow of EAI, and Fellow of BCS. He is the Senior Member of IEEE, IES, and ACM. He is the Distinguished Speaker of ACM. He was included in Most Cited Chinese Researchers (Computer Science) by Elsevier from 2014 to 2018. He was 2019, 2021 & 2022 recipient of Clarivate Highly Cited Researcher. He is included in World’s Top 2% Scientist by Stanford University from 2020 to 2022. He won the Emerald Citation of Excellence 2017, Information Fusion 2022 Best Paper Award, etc. His three papers are included in UK Research Excellence Framework (REF) 2021.He has (co)authored over 400 peer-reviewed articles in the journals: Ann Oncol, JACC, JAMA Psychiatry, IJIM, Inf Fus, IEEE TFS, IEEE TII, IEEE TIP, IEEE TMI, IEEE IoTJ, Neural Networks, IEEE TITS, Pattern Recognition, IEEE TGRS, IEEE JBHI, IEEE TCSVT, IEEE TETCI, IEEE TCSS, IEEE JSTARS, IEEE TNSRE, IEEE SJ, ACM TKDD, ACM TOMM, IEEE/ACM TCBB, IEEE TCAS-II, IEEE JTEHM, ACM TMIS, etc. There are more than 60 ESI Highly Cited Papers and 6 ESI Hot Papers in his (co)authored publications.
His citation reached 25924 in Google Scholar (h-index 89) and 14412 in Web of Science (h-index 66). He is the editor of Neural Networks, IEEE TITS, IEEE TCSVT, IEEE JBHI, etc. He has conducted many successful industrial projects and academic grants from NIH, Royal Society, GCRF, EPSRC, MRC, BBSRC, Hope, British Council, Fight for Sight, and NSFC. He has given over 120 invited talks at famous universities and top conferences, including Harvard University, University of Birmingham, University of Sheffield, University of Surrey, Manchester Metropolitan University, De Montfort University, Polish Academy of Sciences, University of Warsaw, Hasselt University, etc. He has served as (Co-)Chair for more than 60 international conferences and workshops (including more than 20 IEEE or ACM conferences). His research outputs have been reported by more than 50 news press, such as Reuters, BBC, Telegraph, Mirror, Physics World, UK Today News, EurekAlert! Science News, India Times, Association of Optometrists (AOP) news, Medical Xpress, etc.
Title: Data Processing Theories and Techniques for Medical Image Analysis
The field of medical image analysis has witnessed remarkable advancements in recent years, largely attributed to the incredible potential of data processing theories and techniques. This talk aims to provide an overview of our group’s advancements of data processing theories in medical image analysis. The talk will begin with an introduction to deep learning and its vital networks, such as convolutional neural network, advanced pooling networks, graph convolutional networks, attention neural networks, weakly supervised networks, vision transformers, etc. We will explore how these neural networks can be tailored and applied to various medical imaging modalities, including magnetic resonance imaging, computed tomography, and histopathology slides. Furthermore, we will discuss the challenges faced in data processing for medical image analysis, such as limited labelled data, class imbalance, and interpretability, and delve into the theories and techniques employed to mitigate these issues.
Prof. Zhe Wang
Department of Computer Science and Engineering,East China University of Science and Technology
Wang Zhe is a Professor, Doctoral Supervisor, Shanghai ShuGuang Scholar, recipient of Shanghai Talent Development Fund, and a nominee for the National Excellent Doctoral Dissertation. He currently serves as Executive Director of the Shanghai Computer Society, and Member of the Artificial Intelligence and Pattern Recognition Committee of the China Computer Society.
As the first author or corresponding author, he has published or had accepted more than 60 academic papers in internationally renowned journals and conferences such as IEEE Transactions on Cybernetics and IEEE Transactions on Neural Networks and Learning Systems, as well as domestic core journals in the past five years. As the first/second principal investigator, he has authorized/applied for 10/32 invention patents (already published) in the past five years.
As the principal investigator, he has led seven scientific and technological innovation projects, including the National Natural Science Foundation of China, the Shanghai Artificial Intelligence Technology Support Special Project, and the Shanghai Education Commission Key Project of Scientific Research and Innovation. His research focuses on multi-view learning theory and application research, addressing the three major challenges of insufficient use of prior information, imbalanced data categories, and poor interpretability of deep networks. The relevant innovative achievements have been applied in the fields of energy data analysis, medical diagnosis, and aerospace and military industries.
Title：Research on distributed few-shot learning and few-shot generation learning
Abstract：Few-shot learning is a field that aims to tackle the challenge of achieving high performance on new data when there is only a limited amount of training data. In recent years, it has made significant progress and shown effective applications in various domains, such as computer vision and natural language processing. This presentation will introduce distributed few-shot learning and few-shot generation from different perspectives. In the field of distributed few-shot learning, we present methods based on distribution estimation and representation topology optimization, which take into account data privacy issues. We also introduce federated learning, which is based on gradient selection and client customization, to provide customized services while ensuring data privacy. In the domain of few-shot generation, we introduce methods based on frequency domain decomposition, prototype estimation, and latent space editing that achieve high-quality image generation. Finally, we discuss and introduce strategies on how to design more effective models, address differences in research problems, and fully exploit limited data to enhance the performance of few-shot learning.
Prof. Haiwu Li
Xiangsihu College of GuangXi Minzu University
HAI-WU LEE (Member, IEEE) received the degree from the Department of Electronic Engineering, Kun Shan University, in 2000, the master’s degree from the Institute of Computers, Communications, and Control, National Taipei University of Technology, in 2003, and the Ph.D. degree from the Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, in 2014.
He is currently AI Academic Leader and Professor with the Department of School of Science and Engineering, Xiangsihu College of Guangxi University for Nationalities. His research interests include the design and application of optimal control systems for biped walking robots, image processing, voice recognitional and intelligence RFID. He is a Reviewer of journals, such as IEEE TRANSACTIONS ON EDUCATION, Array, and IEEE ACCESS etc….