Welcome to Yubing's homepage!
Tong, Yubing (佟雨兵), Ph.D (EE).
Medical Image processing Group (MIPG) at Radiology Department,
University of Pennsylvania,
#602w, 6th floor of Goddard Laboratories Building, 3710 Hamilton Walk, Philadelphia, PA, 19104
Email/Skype: email@example.com ; firstname.lastname@example.org
Related research and working experience:
April 2016~Present _MIPG/UPenn_ (University of Pennsylvania), U.S.A
Research Associate, working on medical image processing and computer vision
Jan 2011~March2016 _MIPG/UPenn_ (University of Pennsylvania), U.S.A
Postdoctoral research fellow, working on medical image processing and computer vision
Postdoctoral research engineer, working on parallel machine learning algorithm for multimedia retrieval
homepage at LIG: http://mrim.imag.fr/yubing.tong/
Postdoctoral research fellow, working on computer vision and visual perception processing and attention model building
Nov 2006~Dec 2008 _Arcsoft Inc,_ Shanghai, China
Video software engineer and project manager, working on video codec optimization
Jul 2004~Oct 2006 _Electrical System Lab. Beihang University,_ Beijing, China
Team leader, workong on the project "video compressing system development on ADI/DSP-Blackfin561".
Jan 2004~Jun 2004 _Lonovo Corp.__ Beijing, China
Intern as software engineer, working on the project of MPEG2 system codec optimization on TI-DSP (DaVinci DM642).
I have over 10 years of experience in designing and developing algorithms and software systems of image processing, computer vision and machine learning. I have attended multiple academic and industrial projects in China, France and United States with solid programming and strong R&D ability. I have in total 56 publications including 28 journal papers and 3 international patent application. My papers are published on Medical Image Analysis, PLON ONE, Medical Physics, Cognitive Computing, SPIE Medical Imaging conference etc.
After obtaining PhD from Beihang University, I joined in Arcsoft Company at Shanghai, China, which a world leading mutlimedia solution provider with headquarter in Silicon Valley, California. I have been a software engineer and then project manager over 2 years in charge of video software development and optimization, especially for the embedded platform (ARM). The industrial experience achieved includes software product development and delivery from request analysis design, codec, delivery and final release version control.
After that, during my work at Grenoble Informatics Lab (LIG) being ranked as A+ in CNRS (Centre National de la Recherche Scientifique in French which means France National Scientific Researcg Center) in France, I designed a saliency map analysis tool and became the first one to use it in the field of image quality assessment with two invited papers for international conferences of EUVIP2010 and VCIP2010. I also developed parallel video indexing algorithm oriented large scale imbalanced data and first proposed an incremental active learning and multiple classifier algorithm which was also implemented on Grid5000 (France National Grid Platform) with thousands of CPU cores.
With my work on medical image processing at Medical Image Processing Group of University of Pennsylvania from January 2011, I have illustrated the novel automatic anatomic recognition (AAR) approach on upper airway segmentation for female patients with Polycystic Ovary Syndrome (PCOS). The method can segment object much more efficient than that from manual segmentation (seconds vs. hours for one organ). The method can work well on 3D static images and also 4D dynamic images. I also took the study a step forward and proposed a novel fat quantification approach under the supervision from Prof. Jayaram K. Udupa at University of Pennsylvania. A novel approach was proposed to find the optimal slice at which subcutaneous adipose tissue (SAT) area (2D) would be maximally correlated with SAT volume (3D). It can get high consistency of anatomic localization for all subjects which is the main problem in traditional clinical practice. An invent patent for this approach is also being applied.
I have served as a reviewer for more than 20 important peer-reviewed scientific journals including IEEE trans. Medical imaging, Medical Physics, IEEE trans. Signal processing, IEEE trans. Image processing, IEEE trans. Multimedia, Medical Image Analysis, Journal of Computer Vision and Image Understanding (CVIU) et al. I am also a conference reviewer and program commit member for international conferences, IEEE International Conference on Systems, Man, and Cybernetics (2013/2014); IEEE International Conference on Computer Vision Theory and Applications (2014 -), MICCAI 2015. Most of this experience has directly related with the challenges encountered in future projects of medical image processing, computer vision and machine learning techniques.
Medical Image Processing & Analysis: Quantitative Radiolgoy research including medical image segmentation and registration, focus on automatic anatomy recognition approach for organs segmentation in the whole body wide; especially 3D/4D segmentation of upper airway organ system, lymph zones and lymph nodes detection, 3D/ 4D organs deformation and motion research as well as 4D MR image construction from 2D slice acquisitions; fat quantification based standardized anatomy space approach, MRI/CT image intensity standardization and non-uniformity correction, PET-CTquantitative research.
Visual perception and image retrieval with machine learning techniques: visual perception mechanism research as well as gaze/attention model building and model's application, such as image quality analysis with saliency detection models; (active) machine learning techniques in image retrieval with multiple classifiers for large scale un-balanced data set and machine learning techniques for image quality analysis.
Video compress and embedded system application & development: Video compression algorithms / video surveillance system based on H.264, MPEG-4 and HEVC, as well as video codec (encoder and decoder) optimization on the embedded system (DSP or ARM system); application development on Android platform (with Samsung Galaxy S4, S6).
Mandarin (native), fluent in English and Cantonese .