Prof. Tae-Seong Kim
Kyung Hee University, Republic of Korea
Director, Bioimaging and Brain Engineering Laboratory, Kyung Hee University, Republic of Korea
 

Tae-Seong Kim received the B.S. degree in Biomedical Engineering from the University of Southern California (USC) in 1991, M.S. degrees in Biomedical and Electrical Engineering from USC in 1993 and 1998 respectively, and Ph.D. in Biomedical Engineering from USC in 1999. After his postdoctoral work in Cognitive Sciences at the University of California at Irvine in 2000, he joined the Alfred E. Mann Institute for Biomedical Engineering and Dept. of Biomedical Engineering at USC as Research Scientist and Research Assistant Professor. In 2004, he moved to Kyung Hee University in Republic of Korea where he is currently Professor in the Department of Biomedical Engineering. His research interests have spanned various areas of biomedical imaging, bioelectromagnetism, neural engineering, and assistive lifecare technologies. Dr. Kim has been developing novel methodologies in the fields of signal and image processing, pattern classification, machine learning, and artificial intelligence. Lately Dr. Kim has started novel projects in the developments of smart robotics and machine vision with deep learning methodologies. Dr. Kim has published more than 350 papers and ten international book chapters. He holds ten international and domestic patents and has received ten best paper awards.

Speech Title: "Deep Learning AI Methodologies and Their Applications in Biomedical Technologies"

Abstract: In the era of artificial intelligence (AI), biomedical technologies are being transformed into a new domain of biomedical AI. As AI has wide applications in the field of biomedical engineering and technologies, it will transform medicine and healthcare in the near future. Among various machine learning principles and techniques, deep learning is leading this new development of biomedical AI. In this presentation, major deep learning principles and methodologies including convolutional neural networks, recurrent neural networks, auto-encoders, and reinforcement learning will be introduced. Then their AI applications to biomedical technologies will be presented including biomedical computer-aided diagnostic (CAD) systems, human activity recognition of assistive lifecare systems, biomedical machine vision systems, and humanoid robotics.

Prof. Hiroshi Noborio
Osaka Electro-Communication University, Japan
 

Hiroshi Noborio received the B.Eng. and M.Eng. in Department of Computer Science from Shizuoka University in 1982 and 1984, respectively. In succession, he received the Ph.D. in Department of Mechanical Engineering, Faculty of Engineering Science, Osaka University in 1987. Then, he moved to the Osaka Electro-Communication University in Precision Engineering, Engineering Informatics, and Computer Science departments. Also, he worked in TU Munich supported by the Humboldt-Fellowship as a guest researcher. His research interests have spanned various areas of Robotics, CV (computer vision), CG (computer graphics), XR (VR+MR+AR). Prof. Noborio has developed the sensor-based and model-based navigation. Lately he has started to research surgical guiding of dental implant and surgical simulation/navigation of Liver, Kidney and Brain. They were mainly supported by Grants-in-Aid for Scientific Research program established by MEXT and JSPS and so on. Dr. Noborio nominated his research "Interference Check Algorithm Based on the Representation" as one of works in "The History of Robot Research and Development in Japan" as the 20th anniversary of RSJ. Also, he published more than 200 papers and 35 international book chapters, and in addition he has received best paper awards several times.

Speech Title: "Depth-Depth Matching of Virtual and Real Images for a Surgical Navigation System"

Abstract: The key idea of our surgical navigation system is the depth-depth matching (DDM) of virtual and real organ images. The depth image of virtual organ comes from Z-buffer of GPU (Graphics Processing Unit) for a virtual organ modeled by STL (Stereolithography) data. On the other hand, a depth image of real organ comes from some depth image by an arbitrary depth camera for a real organ. Therefore, in DDM, we need only non-combinatorial L subtractions and additions between virtual and real 2D depth images whose pixel number is to be L. L is about hundred thousand. On the other hand, the most popular Iterative Closest Point (ICP) algorithm in Point Cloud Library is time consuming for checking the coincidence of two kinds of point clouds of whole organs. The reason are as follows: (1) The ICP needs combinatorial M*N calculation of the Euclidean distances of 3D cloud points (M and N are usually near hundred thousand). (2) Since a real organ is obstructed by its patient’s body, a captured direction is restricted as the top view on or near the shadow-less lamp.