Keynote Speakers


Keynote Speaker I

Prof. Tuan D. Pham
Prince Mohammad Bin Fahd University, Saudi Arabia


Tuan D. Pham currently holds positions as Professor in AI and Founding Director of the Center for Artificial Intelligence at Prince Mohammad Bin Fahd University, Saudi Arabia. The Center for Artificial Intelligence is equipped with state-of-the-art computing facilities and infrastructure. His previous position was Professor of Biomedical Engineering at Linkoping University, University Hospital Campus, Linkoping, Sweden. He was appointed as Professor and Leader of the Aizu Research Cluster for Medical Engineering and Informatics, and the Medical Image Processing Lab, both at the University of Aizu, Japan. Before his appointments in Japan, he was appointed as Associate Professor and the Bioinformatics Research Group Leader at the University of New South Wales, Canberra, Australia. His current research focuses on AI and machine learning methods for image processing, time-series analysis, complex networks, and pattern recognition applied to medicine, biology, and mental health. He serves as an Associate/Section Editor for several scholarly journals, series, and conference proceedings. In 2020, Dr. Pham is selected as an Expert in Artificial Intelligence by the U.S. Food and Drug Administration (FDA) Center for Devices and Radiological Health (CDRH) Network of Digital Health Experts Program (NoDEx).


Speech Title: "Recurrence Analysis in Deep Learning of Medical Images and Physiological Signals"


Abstract: Time-frequency and time-space properties of medical images and physiological time series are introduced as a robust tool for deep learning-based classification. Experimental results obtained from the classification of benign and malignant mediastinal lymph nodes in lung cancer on computed tomography and sensor-induced physiological signals of Parkinson's disease and heart irregularity show the effectiveness of the use of recurrence analysis in deep learning. The proposed approach has the potential for 1) achieving very high classification accuracy, 2) saving tremendous time for data learning, and 3) being cost-effective and user-comfortable for clinical trials by reducing multiple wearable sensors for data recording.


Keynote Speaker II

Prof. Taesung Park
Seoul National University, South Korea


Prof. Taesung Park received his B.S. and M.S. degrees in Statistics from Seoul National University (SNU), Korea in 1984 and 1986, respectively and received his Ph.D. degree in Biostatistics from the University of Michigan in 1990. From Aug. 1991 to Aug. 1992, he worked as a visiting scientist at the NIH, USA. From Sep. 2002 to Aug. 2003, he was a visiting professor at the University of Pittsburgh. From Sep. 2009 to Aug. 2010, he was a visiting professor in Department of Biostatistics at the University of Washington. From Sep. 1999 to Sep. 2001, he worked as an associate professor in Department of Statistics at SNU. Since Oct. 2001 he worked as a professor and currently the Director of the Bioinformatics and Biostatistics Lab. at SNU. He served as the chair of the bioinformatics Program from Apr. 2005 to Mar. 2008, and the chair of Department of Statistics of SNU from Sep. 2007 and Aug. 2009. He has served editorial board members and associate editors for the international journals including Genetic Epidemiology, Computational Statistics and Data Analysis, Biometrical Journal, and International journal of Data Mining and Bioinformatics. His research areas include microarray data analysis, GWAS, gene-gene interaction analysis, and statistical genetics.


Speech Title: "Deep Learning-Based Pathway Analysis using Hierarchical Structural Component Models"


Abstract: Many statistical methods for pathway analysis have been used to identify key genes and proteins within previously known pathways associated with a given disease, or to detect novel pathways from proteins known to be disease-related. It is well known that pathways overlap and are highly correlated. However, most pathway methods do not consider this correlation between pathways, which may cause false negative or positive errors. Recently, a hierarchical structural component model (HisCoM) was proposed to take this correlation into account by fitting all pathways in one model simultaneously. However, since HisCoM assumes linear contributions of biological factors to the effect of a pathway, it does not fully project the complexity of relationships between biological factors. We propose DeepHisCoM which uses deep learning to find complex contributions of biological factors to the effect of a pathway. DeepHisCoM has all advantages of HisCoM such as using the hierarchical structured information in pathways. Through simulation studies, DeepHisCoM was shown to have higher power in the non-linear pathway effect and comparable power for the linear pathway effect, when compared to the conventional pathway methods. Application to a Hepatocellular Carcinoma (HCC) dataset demonstrated that DeepHisCoM successfully identified well known pathways that are highly associated with HCC such as Tryptophan metabolism and Primary bile acid biosynthesis.


Keynote Speaker III

Prof. Fabio Roli, IEEE&IAPR Fellow
University of Cagliari, Italy


Fabio Roli is a Full Professor of Computer Science at the University of Cagliari, Italy, and Director of the Pattern Recognition and Applications laboratory ( He is partner and R&D manager of the company Pluribus One that he co-founded ( ). He has been doing research on the design of pattern recognition and machine learning systems for thirty years. His current h-index is 71 according to Google Scholar (April 2021). He has been appointed Fellow of the IEEE and Fellow of the International Association for Pattern Recognition. He was a member of NATO advisory panel for Information and Communications Security, NATO Science for Peace and Security (2008 – 2011). Prof. Roli is the recipient of the 2020 Pattern Recognition Medal of the international scientific journal Pattern Recognition, and the 2020 IAPR Pierre Devijver Award, granted to an outstanding scientist who has significantly contributed to the field of statistical pattern recognition.


Speech Title: "From Known Knowns to Unknown Unknowns in AI: Historical and Technical Issues"


Abstract: AI has been originally developed for closed-world, and noise-free, problems where the possible states of natures and actions that a rationale agent could implement were perfectly known. One could argue that, at that time, AI dealt with known knowns. Since the 1980s, when machine learning became an experimental science, AI researchers started to tackle pattern recognition problems with noisy data, using probability theory to model uncertainty and decision theory to minimize the risk of wrong actions. This was the era of known unknowns, characterized by the rise of benchmark data sets, larger and larger year after year, and the belief that real world problems can be solved collecting enough training data. However, recent results have shown that available data sets have often a limited utility when used to train pattern recognition algorithms that will be deployed in the real world. The reason is that modern machine learning has often to face with unknown unknowns. When learning systems are deployed in adversarial environments in the open world, they can misclassify (with high-confidence) never-before-seen inputs that are largely different from known training data. Unknown unknowns are the real threat in many security problems (e.g., zero-day attacks in computer security). In this talk, I give a historical and technical overview of the evolution of AI and machine learning for pattern recognition, and discuss how this evolution can be regarded as a transition from known knowns to unknown unknowns, and the key role that adversarial machine learning plays to make AI safer.










© Copyright ICBBT 2021 | All Rights Reserved