AI Session
SYNAPSE 2026 Conference, 2pm-4pm, 4 June 2026
Shaw Foundation Alumni House (SFAH) Auditorium (Level 2)
Wilson GOH
Biodata
Dr. Wilson Goh is the Chief Data Scientist and Deputy Director at the Centre of AI in Medicine (C-AIM), Nanyang Technological University. Wilson is also a Senior Lecturer (Associate Professor) at Imperial College London and Senior Research Scientist in the Institute of Mental Health, Singapore.
Dr. Goh’s research focuses on leveraging complex multimodal clinical datasets to develop and deploy AI solutions within hospital environments, with particular emphasis on mental health and cancer detection. His recent work has expanded into implementation science, where he explores data governance issues and investigates how clinicians trust and interact with AI systems. Through his research, Dr. Goh aims to bridge the gap between basic science, AI innovation and integration into healthcare
Abstract – Better Mental Health, Better Brain Health: Building Capacity for Early Risk Detection at the Institute of Mental Health, Singapore
Mental and brain health are deeply intertwined, yet services and research often treat them separately. At the Institute of Mental Health, Singapore, we are building clinically grounded, multimodal prediction systems that connect symptoms, cognition, socio-demographics, omics, imaging, digital and language-derived signals into actionable models, improving current outcomes and safeguarding long-term brain health.
Over the last 20 years, we have developed deep cohorts on youths, adults and elderly, combining psychometrics, speech tasks, cognition and socioenvironmental risk modules into clinically practical prediction frameworks that are logistically feasible and externally testable. These cohorts reveal novel and exciting insights. For example, brief speech samples from youths yielded interpretable sentiment and linguistic markers (e.g. diminished sentiment variability, simpler morphology, reduced lexical richness) parallelizing earlier speech work in dementia. Blood-plasma proteomic signatures forecasted transition to psychosis with AUCs up to 0.96 and showed functional convergence with Caucasian-derived signatures in complement, coagulation and lipid pathways. Multimodal analysis revealed that behavior, psychological, and molecular modalities are complementary. The advent of large language models provides unprecedented opportunities for data integration and explainability but requires workarounds on limitations especially on privacy and trust. To better effect care management, we have built the HOPEs cohort to better manage home-based care. Finally, we are moving into population cohorts using lifelong electronic medical records (EMRs) to promote and preserve society mental well-being. These efforts together, form a pragmatic roadmap for embedding early risk detection and management into societal mental health care system to support better lifelong brain health, especially in later life.
Helen ZHOU
Biodata
Associate Professor Juan Helen Zhou is Director of the Centre for Translational MR Research and an Associate Professor at the Centre for Sleep and Cognition, Healthy Longevity Translational Program, Yong Loo Lin School of Medicine, National University of Singapore, with joint appointments in Electrical and Computer Engineering at NUS and affiliations with Duke‑NUS Medical School. Her research investigates selective brain network vulnerability in ageing and neuropsychiatric disorders, using multimodal neuroimaging and machine learning to advance precision neurology and psychiatry. She is internationally recognised for pioneering work in multimodal brain connectomics and brain foundation models. She serves on editorial or advisory boards for leading journals, including Nature Communications Biology, eLife, Human Brain Mapping, and Imaging Neuroscience, and has secured competitive funding from Singapore’s NMRC, NRF, and MOE, as well as the UK Royal Society and the US NIH. She is an OHBM Fellow and contributes to major scientific societies and conferences, including ISMRM, OHBM, NeurIPS, MICCAI, and KDD, through Council, Program/Organization Committee, and Area Chair roles.
Talk Title: Brain Imaging Meets AI: Advancing Precision Neurology and Psychiatry
An BO
Biodata
Bo An is a President’s Chair Professor at Nanyang Technological University, Singapore. He was elected a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI). Dr. An was the recipient of the 2010 IFAAMAS Victor Lesser Distinguished Dissertation Award, an Operational Excellence Award from the Commander, First Coast Guard District of the United States, the 2012 INFORMS Daniel H. Wagner Prize for Excellence in Operations Research Practice, 2018 Nanyang Research Award (Young Investigator), and 2022 Nanyang Research Award. He was invited to give Early Career Spotlight talk at IJCAI’17. He led the team HogRider which won the 2017 Microsoft Collaborative AI Challenge. He was named to IEEE Intelligent Systems’ “AI’s 10 to Watch” list for 2018. He was PC Co-Chair of AAMAS’20 and General Co-Chair of AAMAS’23. He is on the IJCAI Board of Trustees and will be Program Chair of IJCAI’27. He was elected to the Executive Council of AAAI, the board of directors of IFAAMAS, and Distinguished member of ACM.
Abstract – Agentic Reinforcement Learning
LLM-based autonomous agents are playing an increasingly important role in many fields, including general computer control, software engineering, scientific discovery, and social simulation. Early implementations of autonomous agents largely relied on manually constructed workflows based on prompt engineering. However, such static workflows often struggle to generalize effectively to out-of-domain tasks, lack the ability to maintain high-quality, continuous interaction with environments and users, and are unable to achieve self-improvement during long-term operation. As a result, a growing body of research has begun to introduce reinforcement learning, enabling agents to continuously optimize their policies through interaction and thereby systematically improve their generalization, interaction, and tool-use capabilities. This talk will discuss major challenges encountered in agentic RL and some recent progress.
Lim Soon WONG
Biodata
Wong Limsoon is a Fellow of the ACM and the Singapore National Academy of Science and a professor with decades of experience at the intersection of AI, analytics, and scientific reasoning. His research focuses on looking for deep, unchanging truths in data rather than getting distracted by surface-level noise. Whether he is challenging top AI models with his son’s school math puzzles or exposing blind spots in classic technology, Limsoon is interested to see AI push past simple pattern-matching toward genuine, human-like insight.
Abstract – The Gap between Human Intelligence and Artificial Intelligence
While frontier AI excels at absorbing rules and recognizing complex patterns, it lacks the fluid human capability to infer underlying principles when those rules break or no longer apply. This presentation highlights the critical cognitive weaknesses causing this gap, showing how current models struggle with novel reasoning tasks and blindly inherit deep conceptual flaws. While adopting ultra-sparse, brain-inspired network topologies directly addresses AI’s severe energy inefficiencies , its relationship to generating genuine insight remains an open, compelling frontier. Because the biological brain naturally possesses this capacity for insight, researchers must closely study and mirror its structural architecture to build the next generation of truly intelligent AI systems.