T U T O R I A L S
Prof. Vangelis Metsis
Texas State University, USA
Title: Modern methods and tools for human biosignal analysis
Abstract: The term biosignal refers to any signal that can be measured from living organisms. Biosignals have been used in medicine, sports science, and psychology for diagnoses, and there have been impressive advancements in these areas. Recently, the fields of human-computer interaction and affective computing have found an interest in using biosignals as a means of understanding the human state and intention. This interest has been reinforced by the fact that acquiring information with sensors and interfacing electrically with the human body has become much easier in the past few years. Moving from large analog technologies to digital ones has led to the miniaturization of sensing devices. Wireless transmission technologies (e.g., Bluetooth low energy), which can be easily integrated with the acquisition hardware, have removed the need for bulky wiring. This tutorial will present an overview of modern applications of human biosignals and will provide practical examples of machine learning-based methods and tools for biosignal analysis. Traditional machine learning algorithms for feature extraction and classification will be compared with recent developments in deep learning and its applications to biosignal and time-series data processing in general.
Dr. Metsis received his Bachelor of Science degree in Computer Science in 2005, from the Department of Informatics of Athens University of Economics and Business in Greece, and his Doctoral degree in 2011 from the Department of Computer Science and Engineering of The University of Texas at Arlington.
During 2006-2007, Dr. Metsis worked as a Research Associate at the Department of Informatics and Telecommunications of the National Center for Scientific Research (NCSR) “Demokritos” in Greece, contributing to the project MedIEQ, funded by the European Commission. After receiving his Ph.D. diploma, and until joining TxState, he was employed, as a Research Assistant Professor by UTA, and he continued to be affiliated with Heracleia Human-Centered Computing Laboratory, where he was involved in several federally-funded research projects, as a Co-PI or Senior Researcher. He also taught a number of graduate and undergraduate classes at the CSE department.
Dr. Metsis research interests span the areas of Machine Learning, Data Mining and Computer Vision with focus in applications of Smart Health and Wellbeing, Assisted Living and Bioinformatics.
Utilizing Field Programmable Gate Arrays (FPGA) for AI acceleration without noticing it !
Prof. Yannis Papaefstathiou
Aristotle University of Thessaloniki, Greece
Three years ago, in AIAI I have presented an overview of how the designers can utilize FPGAs in their embedded systems, through the use of High-Level-Synthesis (HLS) Tools. In this tutorial we will dive into the new development approaches that allow the designer to take full advantage of FPGAs, both in the Cloud and on the Edge, while barely noticing that their core processing is executed on reconfigurable logic. The new emerging design flow is based on seamlessly utilizing open-source accelerated libraries that are being optimized for execution on the new highly heterogeneous FPGA-based systems.
Ioannis Papaefstathiou is an Associate Professor at the School of Electrical and Computer Engineering at Aristotle University of Thessaloniki and a co-founder of Exascale Performance Systems (EXAPSYS) which is a spin-off of Technical University of Crete and Foundation of Research and Technology Hellas (FORTH). From 2004-2018 he was a Professor at ECE School at Technical University of Crete and a Manager at Synelixis Solutions SA. He is working in the design and implementation methodologies for CPS with tightly coupled design parameters and highly constrained resources as well as in heterogeneous High Performance Computing (HPC) systems and the associated programming/development tools. He was granted a PhD in computer science at the University of Cambridge in 2001, an M.Sc. (Ranked 1st) from Harvard University in 1996 and a B.Sc. (Ranked 2nd) from the University of Crete in 1996. He has published more than 100 papers in IEEE and ACM-sponsored journals and conferences. He has participated in numerous European R&D Programmes(e.g. OSMOSIS, FASTCUDA, HEAP, FASTER, COSSIM ECOSCALE, EXTRA); in total he has been Principal Investigator in 12 competitively funded research projects in Europe (in 7 of them he was the technical manager), in the last 7 years, where his cumulative budget share exceeds €5 million.
Anomaly Detection in Images
Prof. Giacomo Boracchi
Politecnico di Milano, Italy
Anomaly detection problems are ubiquitous in engineering: the prompt detection of anomalies is often a primary concern, since these might provide precious information for understanding the dynamics of a monitored process and for activating suitable countermeasures. In fact, anomalies are typically the most informative regions in an image (e.g., defects in images used for quality control). Not surprisingly, anomaly detection problems have been widely investigated in the image processing and pattern recognition communities, and are key in application scenarios ranging from quality inspection to health monitoring.
The tutorial presents a rigorous formulation of the anomaly-detection problem that fits with many imaging scenarios and applications. The tutorial describes, by means of illustrative examples, the most important anomaly-detection approaches in the literature, and their connection with the machine-learning perspective of semi-supervised and unsupervised learning/monitoring. Special emphasis will be given to anomaly-detection methods based on learned models, which are often adopted to handle images and signals. In particular, these will be divided into traditional models (including dictionaries yielding sparse representations) and deep learning models. The tutorial is accompanied by various examples from our research projects where we applied anomaly-detection algorithms to solve real world problems: visual quality inspection for monitoring chip and nanofiber production.
Giacomo Boracchi is an Associate Professor of Computer Engineering at Dipartimento di Elettronica, Informazione e Bioingegneria of the Politecnico di Milano, where he also received the Ph.D. in information technology (2008), after graduating in Mathematics (Università Statale di Milano, 2004). His research interests concern image processing and machine learning, and in particular image restoration and analysis, change/anomaly detection, domain adaptation. Since 2015 he is leading industrial research projects concerning algorithms for X-ray inspection systems for airport security, automatic quality inspection systems for monitoring silicon wafer production (the system developed with STMicroelectronics is currently analyzing wafer production over different sites), and outlier detection in web-sessions.He is currently associate editor for IEEE Transactions on Image Processing and server as AE for IEEE Computational Intelligence Magazine and in a few special issues. In 2015 he received an IBM Faculty Award, in 2016 the IEEE Transactions on Neural Networks and Learning Systems Outstanding Paper Award and in 2017 the Nokia Visiting Professor Scholarship. He has held tutorials in major IEEE conferences: ICIP 2020, ICASSP 2018 and IJCNN 2017, 2018.