Welcome to 2026 the 10th International Conference on Big Data and Internet of Things (BDIOT2026)
∷
∷
∷
Biosketch: Simon James Fong graduated from La Trobe University in Australia with a First Class Honours BEng degree in Computer Systems and a PhD in Computer Science in 1993 and 1998, respectively. He is currently an Associate Professor in the Department of Computer Science at the University of Macau and a Senior Visiting Scholar at Tsinghua University, Beijing. Prior to joining the University of Macau, he served as an Assistant Professor in the School of Computer Engineering at Nanyang Technological University, Singapore. Dr. Fong has published over 500 peer‑reviewed papers, including top‑tier works in ICDM, ICML, KDD, WWW, ACM MM, MICCAI and other flagship venues in medical analytics and AI. He has been recognized in the Stanford Top 2% Scientists global ranking and has served as a Golden Reviewer for ICML, reflecting his standing in the machine learning research community. He actively serves as a member for IEEE and as Editor‑in‑Chief of the Medical Data Mining journal.
Speech Title: Privacy-Preserving Federated Medical AI: From Local Data to Global Intelligence
Abstract: This talk presents a privacy-preserving federated medical AI framework designed for cross-institutional and cross-regional collaboration without requiring sensitive patient data to leave their local sites. Instead of transferring raw medical images and records, the framework supports local feature learning, synthetic data generation, and incremental model adaptation, while exchanging only controlled model knowledge or meta-features across the federated network. This enables AI models to continue learning from new clinical environments while reducing privacy and data-governance risks. The framework further integrates explainable AI and Legal XAI to provide interpretable evidence, compliance guidance, warnings, and auditable records, helping constrain unreliable AI-generated recommendations and support safer clinical use. Through applications in medical diagnosis, health monitoring, and AI-assisted drug discovery, the talk explores how federated learning can enable international collaboration—including deployment with European partners—under the principle of sharing intelligence rather than sharing private data.
Biosketch: Professor Alexandre Lobo is Vice-Rector for Research and Innovation and Full Professor at the University of Saint Joseph (USJ), Macao SAR, China. He is the founder of the Laboratory of Applied Neurosciences (LAN/USJ) and President of the Board of Directors of the Macau Spin Innovation Incubator, where he supports the translation of academic research into technology-based entrepreneurship and strengthens connections between Macao, Mainland China, the Greater Bay Area, and Portuguese-speaking countries. He holds a PhD in Engineering and a Master’s degree in Artificial Intelligence from the Federal University of Ceará, Brazil, and has an interdisciplinary academic and professional background spanning artificial intelligence, bioengineering, biomedical signal processing, neuroscience, software systems, and project management. His research focuses particularly on the integration of AI with physiological and neurophysiological data, including EEG, ECG/HRV, fNIRS, biosensors, multimodal data analysis, and intelligent systems for health and human performance. Professor Lobo is also Chief Academic Advisor of the National Research Base of Intelligent Manufacturing Services at Chongqing Technology and Business University and has maintained research collaborations with major institutions in China and internationally. He has served as a Visiting Associate Professor at the Shenzhen Institutes of Advanced Technology of the Chinese Academy of Sciences and as an Honorary Research Fellow at the University of Leicester in the United Kingdom. His current work explores longitudinal multimodal neurodata, artificial intelligence, personalized medicine, neurotechnology, and data-driven models of human health and cognition. Alongside his scientific activities, he is actively involved in research strategy, innovation ecosystems, university-industry collaboration, startup development, and international academic cooperation.
Speech Title: Personalized Neurodata Framework: Integrating Longitudinal Data, IoT, and Multiomics for Personalized Medicine
Abstract: Advances in Big Data, Internet of Things, artificial intelligence, and multiomics are creating new possibilities for personalized medicine in neuroscience. However, the main challenge is no longer simply collecting more data, but integrating heterogeneous information across time into clinically meaningful individual profiles. This keynote introduces the concept of a Personalized Neurodata Framework, a longitudinal architecture designed to combine neurophysiological, physiological, behavioral, clinical, environmental, and molecular information. Data from EEG, fNIRS, ECG/HRV, wearable sensors, sleep and activity monitoring, cognitive assessments, and clinical records can be progressively integrated with multiomics layers such as genomics, transcriptomics, proteomics, metabolomics, and microbiome data. The central objective is to move from isolated measurements toward dynamic individual trajectories. Rather than relying only on comparisons with population averages, longitudinal models can establish personalized baselines, detect meaningful deviations, identify patient subgroups, and support prediction of disease progression or treatment response. Artificial intelligence can contribute through multimodal data fusion, representation learning, temporal modeling, clustering, predictive analytics, and explainable decision-support systems. Yet major challenges remain, including interoperability, different temporal resolutions, missing and noisy data, high-dimensional multiomics datasets, privacy, interpretability, and clinical validation. The Personalized Neurodata Framework therefore represents a shift from data accumulation to integrated longitudinal understanding, providing a foundation for more adaptive, predictive, and personalized approaches to neurological and mental health care.
Biosketch: Dr. Wang received his B.S. and Ph.D. degrees in Electronics and Information Engineering from Xidian University, Xi'an, China, in 2004 and 2009, respectively. He is currently a Full Professor at the School of Communication and Information Engineering, Shanghai University, China, and a Senior Member of IEEE and the Chinese Institute of Electronics. He also serves as a technical project evaluation expert for the Shanghai Municipal Government and the National Natural Science Foundation of China. He received the Baosteel Outstanding Teacher Award in 2017. His research interests include the Internet of Things, intelligent information processing, and pattern recognition. To date, he has published seven books and more than 130 scientific papers.
Speech Title: Big Data and IoT for Intelligent Transportation: Advances in Rail Transit Channel Characterization and AI-Driven Urban Traffic Risk Prediction
Abstract: Next-generation intelligent transportation systems (ITS) rely on two indispensable pillars: the Internet of Things (IoT) providing reliable wireless connectivity as the physical foundation, and big data analytics driving intelligent decision-making as the functional core. This keynote presents our systematic research progress in two complementary domains: rail transit wireless channel modeling, and AI-powered urban traffic flow forecasting and risk early warning at the decision layer.
The characteristics of non-stationary massive MIMO millimeter-wave channels are investigated in typical rail transit scenarios, including reinforced concrete tunnels, viaducts, mountainous railways, and ultra-high-speed maglev lines. Improved geometry-based stochastic channel models are proposed, which integrate heterogeneous scatterer complexity, periodic pole scattering effects, and the birth–death dynamics of multipath components, thereby reducing channel prediction errors.
For urban road traffic networks, a multi-view graph convolutional recurrent network is proposed, which jointly models geographic connectivity, semantic traffic pattern similarity, and temporal traffic dynamics, achieving accurate multi-step traffic flow forecasting. In addition, a multi-branch risk prediction model is further developed with topology-adaptive graph convolution, fusing road structure, area characteristics, spatiotemporal traffic activity, and external environmental factors to enable segment-level accident risk estimation.