---

Abstract

Computational museology is a scaffold that unites machine intelligence with data curation, ontology with visualization, and communities of publics and practitioners with embodied participation through kinaesthetic interfaces. Computational museology also empowers cultural organisations to link all forms of culture and materiality: objects, knowledge systems, representation and participation. Research at the Laboratory for Experimental Museology (eM+) reaches beyond object-oriented curation to blend experimental curatorship with contemporary aesthetics, digital humanism and emerging technologies. This lecture explores key themes including interactive archives and emergent narrative, deep fakes, generative AI and blockchain sovereignties, embodied knowledge systems and performative interfaces and scientific visualization for museums in the age of experience.

---

Biography

Professor Sarah Kenderdine researches at the forefront of interactive and immersive experiences for galleries, libraries, archives and museums. She is professor at the École Polytechnique fédérale de Lausanne (EPFL), Switzerland, leading the Laboratory for Experimental Museology (eM+) and is curator-at-large of EPFL Pavilions. Her upcoming book is Deep Fakes: A Critical Lexicon of Digital Museology published by Routledge (2025). Sarah has pioneered the framework of computational museology, a scaffold that unites machine intelligence with data curation, ontology with visualization, and communities of publics and practitioners with embodied participation through kinaesthetic and interactive interfaces. She has created over 110 exhibitions and major installations. Sarah currently (2024) leads USD10million in research grants.

---

Abstract

This talk explores the fascinating evolution of Artificial Intelligence (AI), tracing its path from Generative AI, agentic AI and to the emerging frontier of Physical AI. We will delve into the transformative impact of Generative AI technologies and its AI blueprint for application development, which have revolutionized various industries and creative processes. The presentation will then shift focus to the cutting-edge field of Generative AI, Agentic AI and Physical AI.

---

Biography

Dr. Charles Cheung is a deep learning solution architect at NVIDIA. Before joining NVIDIA, He was a technical manager in a R&D center to lead the machine vision team to develop different solution such as defect inspection, 3D reconstruction and 3D recognition for different industries. He received his PhD in Hong Kong Baptist University in applied mathematics focusing on numerical analysis for partial differential equation and meshless collocation method.

---

Biography

Professor Johnny Poon is the Associate Vice-President (Interdisciplinary Research) at Hong Kong Baptist University. He oversees the research development of four strategic research clusters and six interdisciplinary laboratories at the university.  Professor Johnny Poon is also the Dr. Hung Hin Shiu Endowed Professor in Music and Founding Dean of the School of Creative Arts. His research focuses on interdisciplinary projects that break traditional artistic and disciplinary boundaries and advance the future of creative practice. He is currently the deputy project coordinator of a 5-year research project, "Building Platform Technologies for Symbiotic Creativity in Hong Kong", funded by the Research Grants Council under the Theme-based Research Scheme. This project intends to harness the power of science and technology to radically advance human and AI interaction in the arts. Of his professional practice, he has maintained a highly successful career as a conductor, appearing regularly in concert halls and music festivals across the globe.

---

Abstract

Visual Language Models (VLMs) have transformed the landscape of AI-generated art, opening new possibilities for creative expression and human-AI collaboration. Our research introduces the Identification-Explanation-Implication (IEI) framework, which draws inspiration from artistic practices and cognitive science to analyze and enhance AI's creative capabilities. By studying artist-generated visual mashups—where elements from different sources are deliberately combined to create new meaning—we investigate how VLMs understand and generate creative combinations. Our findings reveal that while AI systems can effectively comprehend creative combinations, often surpassing casual observers, they still fall short of matching expert artists' nuanced understanding. More importantly, we demonstrate that by explicitly incorporating artists' conceptual blending processes into AI systems' generation pipeline, we can significantly improve their creative outputs. This work offers both theoretical insights for artists working with AI and practical guidelines for developing more sophisticated AI art tools, contributing to the ongoing dialogue between traditional artistic practices and emerging AI capabilities.

---

Biography

Dr. Yixin Zhu is an Assistant Professor and the Assistant Dean at the Institute for Artificial Intelligence, Peking University. He received a Ph.D. degree from UCLA in 2018, advised by Prof. Song-Chun Zhu. His research builds interactive AI by integrating high-level common sense (functionality, affordance, physics, causality, intent) with raw sensory inputs (pixels and haptic signals) to enable richer representation and cognitive reasoning on objects, scenes, shapes, numbers, and agents.

---

Abstract

With the massive amount of computational power brought by graphics processing unit (GPU), deep learning has now become the hottest branch of Artifical Intelligence (AI). However, this is not to say that traditional solutions are no longer necessary. Researchers now explore traditional and modern methods, respectively, to solve many complex vision problems, such as depth estimation, pose regression, etc. In this talk, I will introduce a few selected projects carried out in the UIC 3D Lab. In particular, I will discuss how we solve problems as camera calibration, 3D modelling, re-lighting, and pedestrian re-identification, etc.

---

Biography

Professor Hui Zhang received the PhD degree in computer vision from the University of Hong Kong in 2006. Since then, she has been with the Department of Computer Science at BNU-HKBU United International College, where she is now a professor and the director of 3D Laboratory. Her research interests are in computer vision and machine intelligence. Her research works include camera calibration, 3D reconstruction, relighting, re-ID, etc. She has published over 60 peer-reviewed journal and conference papers. Many of her works appear in top venues, including CVPR, ACM MM, ECCV, ICRA, TPAMI, and TIP. She regularly serves as a reviewer of major computer vision and artificial intelligence conferences (e.g., CVPR, ECCV, ICCV).

---

Abstract

The creation of lifelike digital humans is a critical task in entertainment, education, and the Metaverse, motivated by the demand for immersive virtual experiences. This talk introduces our creative pipelines in addressing the challenge from efficient 3D representation to capture human appearance, AI-driven motion capture for realistic dynamics, and physics-based modeling to ensure natural interactions with virtual environments. Our generative rendering techniques bring these characters to life with photorealistic visuals, enabling real-time and interactive digital humans for various applications.

---

Biography

Dr. Chen Jie is currently an Assistant Professor at the Department of Computer Science, Hong Kong Baptist University. Before joining HKBU, he was a senior algorithm engineer at OmniVision Technologies Inc., Singapore. He worked as post-doctoral research fellow at ST Engineering-NTU Corporate Laboratory, Singapore for three years after he received Ph.D. degree in the School of Electrical & Electronic Engineering, Nanyang Technological University, Singapore. He received B.Sc. and M. Eng. degrees both from School of Optical and Electronic Information, Huazhong University of Science and Technology, China. His research focuses on mathematical and machine learning models that solve computational photography and computer vision problems. Dr. Chen has publications in top-tier journals and conferences (TIP, TCI, TCSVT, TGRS, CVPR, ECCV, etc.), and he has been frequently serving as PC/AC members at international academic venues (ACM MM, ICME, VCIP, EUSIPCO etc.). He currently serves as an Associate Editor for the Visual Computer Journal, Springer.

---

Abstract

The creation of 3D digital assets has recently garnered significant attention. With the widespread adoption of large foundation models and neural representations such as NeRF and 3D Gaussian Splatting (3DGS), generating 3D assets has become increasingly accessible. However, the security of these generated assets is often overlooked. During the distribution of those assets,malicious users can easily steal these assets and use them without authorization, potentially infringing on the ownership rights of the original creators. Furthermore, if these 3D assets are altered or manipulated, it could lead to serious consequences. In light of these challenges, there is a pressing need for effective methods to safeguard the ownership of 3D digital assets and enable data owners to assert their rights. In this talk, we will introduce our work on 3D digital asset protection, focusing on preventing those asset from the potential misuse.

---

Biography

Dr. Wan Renjie received his BEng degree in Network Engineering from the University of Electronic Science and Technology and his Ph.D. degree from the Interdisciplinary Graduate School of Nanyang Technological University in Singapore. His research interests include Computational photography, 3D vision, and AI security. From 2019 to 2020, he was a guest researcher at Peking University. From 2020 to 2021, he was a Wallenberg-NTU Presidential Postdoctoral Fellow supported by Wallenberg Foundation in Sweden and NTU. Dr. Wan has several publications in top journals and conferences (TPAMI, IJCV, TIP, CVPR, and ICCV). He was the recipient of the outstanding reviewer of ICCV 2019 and VCIP 2020 best paper award.

---

Abstract

The concept of Multimodal AI Assistant has gained significant traction within the AI community due to its vast potential across a wide array of real-world applications, ranging from autonomous driving and AR/VR to education and healthcare. A crucial step in developing such an assistant involves aligning vision and language modalities, a task that is both challenging and resource-intensive. In this talk, I will present a straightforward approach to bridging the modality gap by employing scene graphs. Specifically, I will focus on scene graph generation from a dataset perspective, encompassing various types of data such as images, videos, and 3D.

---

Biography

Dr. Kaiyang Zhou is an Assistant Professor in the Department of Computer Science at Hong Kong Baptist University, specializing in machine learning and computer vision. He has published over 30 technical papers in leading venues, including CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, AAAI, and top journals like IEEE TPAMI and IJCV. His work has garnered over 10,000 citations. He serves as an Associate Editor of IJCV, the flagship journal in computer vision, and regularly takes on roles such as Area Chair for major conferences like CVPR, ECCV, NeurIPS, and ICLR. He is also the creator of influential AI software packages, such as Torchreid (the No.1 popular person re-identification software on GitHub).

---

Biography

Professor Simon See is currently the Solution Architecture and Engineering Senior Director and Chief Solution Architect & Global Head for Nvidia AI Technology Center, NVIDIA Corporation. He is also a Professor at Shanghai Jiao Tong University, Professor in Beijing University of Posts and Telecommunications (BUPT), and Professor in Universitas Indonesia (UI). He is being conferred as a Distinguished Fudan Scholar in September 2018 by Fudan University, Shanghai, China. Previously Professor See is also the Chief Scientific Computing Advisor for BGI (China) and has a position in Nanyang Technological University (Singapore) and King-Mong Kung University of Technology (Thailand). Professor See is currently involved in a number of International computational, mathematical science projects and national AI initiatives. Recently Dr Simon has been appointed as the Executive Director of the ASEAN Applied Research Centre (AARC). His research interests are in the area of High-Performance Computing, Big Data, Artificial Intelligence, Machine Learning, Computational Science, Applied Mathematics and Simulation Methodology. Professor See is also leading some of the AI initiatives in the Asia Pacific. He is a Steering Committee member of NSCC’s flagship High Performance Computing Conference Supercomputing Asia (SCA) since March 2018. He has published over 200 papers in these areas and has won various awards.

Professor See is a Fellow member of IET, Chairman of TaskForce of IEEE’s CIS Neural Networks Technical Committee, a member of IEEE, SIAM, AAAI, and also on the Advisory Team of AIP (AI Professional Association), International Advisory Board of Institute of Operations Research & Analytics (IORA), Advisory Team of Machine Intelligence and Data Analytics Research Centre (MIDARC), School of Computing Sciences (India) and Board of Studies, MS Tech Programs of Mahindra University and the committee member of more than 50 conferences.

Professor See graduated from the University of Salford (UK) with a Ph.D. in electrical engineering and numerical analysis in 1993. Prior to joining NVIDIA, Dr. See worked for SGI, DSO National Lab. of Singapore, IBM, International Simulation Ltd (UK), Sun Microsystems and Oracle. He is also providing consultancy to a number of national research and supercomputing centers.

---

Abstract

Chronic liver disease ranks as the 12th leading cause of death worldwide, with significant prevalence in Asia, where chronic hepatitis B, metabolic dysfunction-associated steatotic liver disease, and alcohol-related liver disease affect over 40% of the population. Despite the diverse mechanisms underlying liver injury, these diseases share a common trajectory characterized by fibrosis, which can progress to cirrhosis—a critical juncture associated with heightened risks of liver cancer and severe complications. This talk explores the multifaceted applications of AI in addressing liver diseases, emphasising its potential to enhance existing laboratory tests, imaging modalities, and predictive models tailored to the liver's natural history.

At the population level, AI aims to identify individuals at risk for severe liver disease, facilitating early intervention and treatment. Clinically, AI applications focus on assessing liver inflammation, fibrosis, and portal hypertension, alongside advancing predictive models for liver cancer. While numerous studies have proposed AI solutions, many have relied on static baseline parameters; future research must prioritize dynamic data integration to improve predictive accuracy. Additionally, AI's capacity to analyse liver biopsy slides has been validated, offering reduced variability in assessments and expedited diagnostics. Similarly, AI's role in interpreting imaging studies—particularly for cancer detection—holds promise for transforming liver disease management.

---

Biography

Professor Vincent Wong holds the position of Mok Hing Yiu Professor of Medicine and serves as the head of the Division of Gastroenterology and Hepatology at The Chinese University of Hong Kong. His research primarily revolves around viral hepatitis and steatotic liver disease, with a particular focus on the development and application of non-invasive tests for screening, diagnosis and prognostication. Notably, he contributed to the writing committee responsible for formulating the guidelines of the Asian Pacific Association for the Study of the Liver, American Gastroenterological Association and the European Association for the Study of the Liver. He has authored more than 650 publications in international medical journals. His h-index by Google Scholar is 132 as of December 2024. He holds the position of associate editor for the Journal of Hepatology and serves as an editorial board member for renowned publications such as Gastroenterology, JHEP Reports, and Clinical and Molecular Hepatology. He was the president of the Hong Kong Association for the Study of Liver Diseases from 2015 to 2017. Currently, he serves as the Chairman of the Specialty Board of Gastroenterology and Hepatology at the Hong Kong College of Physicians. He has been honoured with research awards from esteemed organizations such as the Asian Pacific Association for the Study of the Liver, Asian Pacific Digestive Week Federation, British Society of Gastroenterology, Hong Kong College of Physicians, the Health Bureau of the Hong Kong SAR Government, and the Ministry of Science and Technology and the Ministry of Education of China.

---

Abstract

In order to achieve sustainability in tackling preeminent healthcare challenges, such as those in population health (e.g., infectious disease epidemiology), carefully designed and deployed AI agents will play a critical role in engaging (e.g., learning/understanding, and thereafter planning/making decisions for addressing) complex multiscale, multifactorial dynamic systems. Towards this end, several fundamental research questions need be to be responded to that arise from the real world and are concerned with resilience and responsiveness in the AI agents. This talk will examine some of those questions via illustrative examples; namely: What kind of complexity will the AI agents deal with? What kind of learning/cognition capacity will be essential in the AI agents, so that they can sustainably perform, i.e., flexibly (seamlessly), economically (with minimal-cost), and reliably (with computational/mathematical guarantees) act on intrinsically-complex, problem-specific conditions (e.g., complexity in dynamic systems behavior, and limitations in observation), as well as solution-critical constraints (e.g., diminishing learning time and/or energy resources)?

---

Biography

Professor Jiming Liu has been the Chair Professor in Computer Science (04/2010-) and Associate Vice-President (Research) (05/2017-12/2019) at Hong Kong Baptist University. He received his M.Eng. and Ph.D. degrees from McGill University (Centre for Intelligent Machines). His current research interests include Data Analytics and Machine Learning, Complex Networks and Systems, Web Intelligence (WI), Autonomy-Oriented Computing (AOC) Paradigms, and AI-Enabled Epidemiology and Practice. He is a Fellow of the IEEE. Prof. Liu has served as the Editor-in-Chief of Web Intelligence Journal (IOS), and the Associate Editor of IEEE Transactions on Cybernetics, IEEE Transactions on Knowledge and Data Engineering, Data and Information Analytics (AIMS), and Computational Intelligence (Wiley), among others. He was the Chair of IEEE Computer Society Technical Committee on Intelligent Informatics (TCII).

---

Abstract

The Hospital Authority (HA), supporting the Public healthcare of HK citizens, has been established for over 30 years and has recently been accredited to HIMSS Stage 7, gaining international renown. With three decades of high-quality, structured, and standardized clinical data across Hong Kong under the Clinical Management System (CMS), the data asset is invaluable for supporting clinical research and innovation in Hong Kong. Over the past years, the clinical data, encompassing around 6 billion records, has supported over 450 researchers and 70 research projects at local universities.

Looking ahead, the clinical data will further be utilized to support the Chief Executive’s 2021 Policy Address, which aims to develop Hong Kong into an international innovation and technology center. To this end, life sciences and health tech innovation have been identified as Hong Kong’s strategic positioning. One of our core strengths is leveraging the valuable health data for research and development (R&D).

On the HA service front, it is envisioned that leveraging this compiled data will lead to the development of numerous AI solutions to address various strategic service challenges and deliver improvements. Over the past year, many clinical and non-clinical solutions have been deployed and validated, relieving frontline workloads, improving clinical quality, and enhancing operational efficiency.

We are looking forward for the ecosystem collaborations to support clinical research and development along with healthcare advancement.

---

Biography

Mr. Dennis Lee currently serves as the Senior System Manager for Artificial Intelligence Systems of the Hong Kong Hospital Authority. He current work involve developing the Artificial Intelligence and Big Data Platform to streamline data acquisition for facilitating HA data analysis, develop Hospital Command Center dashboards, and solution deployment for Artificial Intelligence. Mr. Lee leads the Corporate Project management office and as program managers for several large scale system implementations under IT&HI division.

---

Abstract

Endoscopic Retrograde Cholangiopancreatography (ERCP) is a critical minimally invasive procedure for diagnosing and treating biliary and pancreatic disorders. Despite its clinical significance, ERCP's complexity and the high level of technical skill required pose significant challenges to its widespread adoption and effective practice. This report will delve into the specific challenges encountered in ERCP, including the difficulty in mastering the technique, the need for precise anatomical knowledge, and the associated risks of complications. We will then outline a range of innovative solutions designed to address these issues, aiming to improve the training of medical professionals and enhance the safety and efficacy of ERCP procedures.

---

Biography

Dr. Yunlu Feng, Associate Professor at the Department of Gastroenterology, Peking Union Medical College Hospital, Master’s Supervisor. Her professional focus includes early gastrointestinal cancer and the diagnosis and treatment of pancreaticobiliary interventions with endoscopy. She is skilled in various endoscopic diagnostic and therapeutic techniques, including gastroscopy, magnification endoscopy, endoscopic ultrasound, and ERCP. Dr. Feng has received advanced training at the University of California, San Francisco, and Florida Hospital Orlando. She participated in an advanced training course for endoscopic ultrasound teachers organized by the Asian Endoscopic Ultrasound Union and obtained certification. Dr. Feng has published more than fifty articles in journals such as Endoscopy, Digestive Endoscopy, Endoscopic Ultrasound, WJG, Pancreas, Pancreatology, and Surgical Endoscopy. She holds five national invention patents and one software copyright. Currently, she serves as a member of the Digestive Endoscopy Committee of the Chinese Medical Association, Secretary of the Endoscopic Ultrasound Group of the Digestive Endoscopy Committee, and a member and Secretary of the Digestive Endoscopy Committee of the Beijing Medical Association.

---

Biography

Professor Shouyan Wang is the Vice Dean Institute of Science and Technology for Brain-Inspired Intelligence, Director of Neuromodulation and Brain-machine-interface Centre, Fudan University. He is the Chairman of Neuromodulation Branch of Chinese Neuroscience Society. Vice Board Chairman of Brain Computer Interface Alliance of China. He had conducted research on neural information decoding and brain stimulation at University of Oxford (2002-2007), University of Southampton (2007-2012) and Chinese Academic of Sciences (2012-2017) before he joined Fudan University. He has developed research platform for neuromodulation, which include research facilities for invasive neuromodulation, non-invasive neuromodulation, advanced technology development of intelligent neuromodulation and neuroimaging as well. His research interests are intelligent neuromodulation, digital healthcare and translational research into clinical applications.

---

Abstract

In recent years, medical artificial intelligence has rapidly developed, continuously exploring innovation, and significantly improving the quality and efficiency of medical services. A plethora of innovative achievements have emerged in key scenes of medical artificial intelligence such as classification, detection, and segmentation. This report summarizes works such as representation learning, capsule networks to enhance classification performance, clinical priors, multimodal modeling to enhance automatic detection capabilities, self-supervised learning, and few-shot learning to achieve more accurate segmentation.

---

Biography

Dr. Hongxia Xu obtained her B.S. and Ph.D. from Zhejiang University. She currently serves as the AI Technology Lead at WeDoctor Group and Associate Researcher at the Realdoctor AI Lab, Zhejiang University. Dr. Xu's research focuses on the intersection of medicine and artificial intelligence and the industrial translation of these technologies, including early cancer detection and precision treatment, medical large language models, and intelligent medical imaging detection and segmentation. In recent years, she has published over 50 papers in international journals and top-tier computer science conferences such as Nature Communications, Biomaterials, Nano Letters, NeurIPS, and ICLR. Dr. Xu holds 13 invention patents, has led one project funded by the National Natural Science Foundation of China, and participated in seven projects funded by the NFSC and the Zhejiang Provincial Natural Science Foundation. She has received prestigious awards, including the Second Prize of the China Institute of Electronics Science and Technology Award (2022) and the Second Prize of the China Industry-University-Research Cooperation Innovation Achievement Award (2022).

---

Abstract

Explainability has increasingly become a core requirement for intelligent medical device software and AI models. Nevertheless, current medical AI technologies suffer from the ‘interpretability gap’, despite tremendous efforts for enhancing explainability. Here we propose class-association manifold learning, a generative AI technology to enhance explanation of black-box medical AI models. We adopt manifold learning techniques to efficiently decouples common decision-related patterns into a low-dimensional subspace against individual backgrounds, while preserving near-perfect diagnostic decision accuracy. The extracted pattern knowledge is further used to enable intentional modifications on arbitrary samples to visualize global diagnosis rules. We carry out experiments on an extensive set of medical AI tasks, showing that our method not only achieves much higher accuracy in explaining the behavior of medical AI models, but also helps with extracting medical-compliant knowledge that are unknown during model training, thus providing a potential means of assisting medical knowledge discovery.

---

Biography

Professor Yunpeng Cai is the executive director of the Research Center for Biomedical Information Technology, Shenzhen Institutes of Advanced Technology (SIAT), Chinese Academy of Sciences. His research interests include bioinformatics, medical AI, health informatics and health big data mining, with specialties in EMR big data mining, metagenomics and multi-omics analyses, graph neural networks and explainable machine learning. He published over 120 papers including those on top journals such as Nucleic Acid Research, Briefings in Bioinformatics, PLOS Computational Biology, IEEE TOC and IEEE. TNNLS. He is currently the director of the Guangdong Engineer Center & Shenzhen Engineer Lab for Health Big Data Analysis, and editorial board member of the Health Information Science and Systems journal.

---

Abstract

In this talk, I would like share a joint project with the Chinese University of Hong Kong (CUHK) on Hepatocellular Carcinoma (liver cancer) prediction based on clinical data. This project was initiated by a Professor of Faculty of Medicine, CUHK. We started using simple and classic machine learning algorithms, including random forest and boosting on a single time-point 46 laboratory parameters, and then extended to time-series laboratory parameters, and recently multimodality (time series and CT images). I will share the experimental results in three different phases using real clinical liver patient data from the Hospital Authority Data Collaboration Laboratory.

---

Biography

Professor Pong C Yuen received his Ph.D. degree in Electrical and Electronic Engineering from The University of Hong Kong. He joined the Hong Kong Baptist University in 1993, served as the Head of Department of Computer Science from 2011 – 2017 and Associate Dean (Teaching and Learning) of the Faculty of Science from 2018-2024. Currently, he is a Chair Professor in Computer Science, Hong Kong Baptist University.

Over the years, Prof. Yuen has spent his sabbatical and visited a number of universities/research institutes as a visiting professor/scholar, including The University of Sydney (Australia), The University of Maryland at College Park (USA), INRIA Rhone Alpes (France), ETH Zurich (Switzerland), and The University of Bologna (Italy). Prof. Yuen was the director of Croucher Advanced Study Institute (ASI) on biometric authentication in 2004 and the director of Croucher ASI on Biometric Security and Privacy in 2007. He has been serving as the Director of IAPR/IEEE Winter School on Biometrics since 2017.

Prof. Yuen has been actively involved in many international conferences and professional community in the capacity of general/program co-chair, including ICPR 2006, BTAS 2012, ISBA 2016, WIFS 2018, IJCB 2021. He served as Associate Editor of IEEE Transactions on Information Forensics and Security (2014 – 2018, 2021-2022), and received the Outstanding Editorial Board Service Award in 2018. Prof. Yuen has served as the Vice President (Technical Activities) of the IEEE Biometrics Council (2017-2019), and Associate Editor/Senior Editor of SPIE Journal of Electronic Imaging (2012-2019). Currently, Prof. Yuen is serving as the Senior Area Editor of IEEE Transactions on Information Forensics and Security, Editorial Board Member of Pattern Recognition, and Associate Editor of IEEE Transactions on Biometrics, Behaviour and Identity Science. He received the first-prize and second-prize Natural Science Awards from the Guangdong Province and the Ministry of Education, China, respectively. He is a Fellow of IAPR.

---

Abstract

AI has been rigorously explored in healthcare in recent years. Most of the AI models for healthcare are learned from the EHR data where data missingness and scarcity is often unavoidable. To achieve clinically accurate AI, it is desirable to properly incorporate prior knowledge for the model learning. In this talk, I will present some recent works of my research group where explicit and implicit knowledge from medical ontologies and LLMs are leveraged to guide the learning for analytic applications like diagnosis prediction and radiology report generation.

---

Biography

Profesoor William K. Cheung received the Ph.D. degree in computer science from the Hong Kong University of Science and Technology in Hong Kong in 1999. He is currently the Professor of the Department of Computer Science, Hong Kong Baptist University, Hong Kong. His current research interests include artificial intelligence, data mining, collaborative information filtering, social network analysis, and healthcare informatics. He has served as the Co-Chairs and Program Committee Members for a number of international conferences on areas including artificial intelligence, machine learning, data mining, Web intelligence, and health informatics. He is the Program Co-chair of 2020 IEEE International Conference on Health Informatics. From 2002-2018, he was on the Editorial Board of the IEEE Intelligent Informatics Bulletin. He is a Track Editor of Web Intelligence Journal and an Associate Editor of Journal of Health Information Research.

---

Abstract

Ultrasound imaging is widely used for its non-invasiveness, portability, and cost-effectiveness. However, mastering freehand ultrasound requires years of training, contributing to a global shortage of skilled sonographers and delays in medical diagnoses. AI offers a promising solution to bridge this expertise gap, improving ultrasound accessibility across diverse clinical and remote settings. In this talk, I will explore how AI can simplify sonography from two perspectives: during examinations, by providing real-time 3D spatial and language-guided assistance, and after examinations, by automatically summarizing key findings to enhance diagnostic efficiency, accuracy, and workflow.

---

Biography

Dr. Xiaoqing Guo is an assistant professor in the Department of Computer Science at Hong Kong Baptist University. She is also a visiting researcher affiliated with the Noble Group at the University of Oxford. She was a postdoctoral researcher in the Department of Engineering Science at the University of Oxford, working on the projects VisualAI and Turing AI WLR Fellowship, during 2023-2024. She obtained her Ph.D. degree in the Department of Electrical Engineering at City University of Hong Kong in 2022 and received her B.Eng. degree from Beihang University in 2018. She was a research assistant in the Department of Electrical Engineering at Tsinghua University during 2017-2018. Her research interest is in the interdisciplinary field of AI and healthcare, aiming to create innovative intelligent systems that can support high-quality human-machine interaction/collaboration and trustworthy clinical decision-making. In AI and medical imaging fields, she has published over 30 top journal and conference papers, including TPAMI, IJCV, CVPR, ICCV, ECCV, TMI, MedIA, MICCAI, and Nature. She has been selected as one of World’s Top 80 Chinese Young Female Scholars in AI, the Rising Star of Women in Engineering by Asian Deans’ Forum 2024, and she has received prestigious awards, such as CVPR Outstanding Reviewer Award, MSRA Fellowship Nomination Award, Outstanding Doctoral Research Award, Outstanding Research Thesis Award, and three year consecutive Outstanding Academic Performance Award.