Shaoting Zhang's Homepage

The project is supported by Natural Science Foundation of China (NSFC), Grant No. 70581171, and the Shanghai Municipal Research Fund, Grant No. 045118045. Intel sponsors the Multi-Core platform as well as the force feedback device, and Shanghai Renji Hospital provides clinical experience. The modeling group consists of 12 students: 4 come from CS of TU Berlin, 2 ACM Class of SJTU, 4 SE, 3CS, and I am the team leader.

This research consists of mathematical models, collision detection, collision response, cutting process and force feedback programming. We focus on the Constrained Particle System, M-Rep and FEM to model the soft tissue. The target is to achieve an effective and efficient deformation effect. We also advanced the Hierarchical Spatial Hashing and Progressive Cutting algorithm to simulate the Surgery process. The PHANToM Desktop is choosen as the force feed-back device, and Intel Multi-Core CPU is employed to accelerate our multi-thread application.

		Collision Detection & Constrained Particle System - *Team Leader* It is a general research concerned with the basic algorithm of virtual surgery. The Constrained Particle System is considered as the deformable model, a novel adaptive spatial partition hash algorithm is invented to detect the collision between deformable objects in real time. We will cooperate with Shanghai Renji Hospital and focus on special clinical cases (virtual surgery in the virtual endoscopy environment) based on these general algorithms in the following semester. (Ubuntu 6.06, C++, Python2.4, wxPython, VTK5.0, 2006.3-present) Demo (VR algorithms) & Demo(Virtual Endoscopy) Thanks to DemoRecorder for providing the screen recorder tool in linux.
		Segmentation Validation - *Team Leader* Validation for image segmentation plays an important role in medical image processing. Point-mappings based validations process segmentation results by matching corresponding points on surfaces of the segmentation model and the standard model, then evaluate the matches locally and globally. We proposes four algorithms of building point-mappings; and creatively applies perfect matching algorithm of weighed bipartite graph on it. Furthermore, detailed local and global evaluation ethods are also employed. (Windows XP, VC7.0 (MFC), 2006.2-2006.10) Demo
		1.5D Barcode Recognition - *Teaching Assistant* The 1.5 dimension barcode has 12 basic shapes, including six triangles, and six rectangles with three different directions and two widths. There is a coding rule existing, according to which different combination of two barcodes represents different number. The recognition processes: 1. Capture the video of the camera. 2. Extract the barcode image. 3. Using scanning lines to analyse the image and decode the barcode. (Windows XP, VC8.0 (MFC) , 2006.7-2006.9 )
		Mass-Spring Model with Centerline - *Core Member* The model is based on an improved elasticity theory using the skeleton structure of the deformable object to reflect volumetric information. The model is refined to satisfy the real-time computation requirement and achieve a reasonable deformation effect on the global level as well as the local region. (Windows XP, VC7.0 (MFC) , VTK4.2, OpenGL, 2004.12-2005.5) Demo
		MS-MREP Model - *Core Member* This model consists of a deformable centerline and a dynamic surface reconstruction mechanism based on a Mass Spring model and Medial Representation respectively. If the external force is relativelylarge, it will be directly transferred to the deformable centerline and the surface will be recreated dynamically according to the position of the centerline using a specific reconstruction mechanism. It works more effectively and efficiently than traditional elastic models on the global level due to employing the Medial Representation to reflect the internal information and the Mass-Spring system to reduce response time. (Windows XP, VC7.0 (MFC), VTK4.2, OpenGL, 2005.3-2005.11) Demo (Solid) & Demo (Cutting)

Selected Publications

Real-Time Virtual Surgery Simulation Employing MM-Model and Adaptive Spatial Hash
Shaoting Zhang , Lixu Gu, Weiming Liang and Jingsi Zhang
16th International Conference on Artificial Reality and Telexistence, ICAT2006 , LNCS 4282, Nov., 2006, pp. 666-675.
Download: pdf [1.01 MB] oral presentation [1.35 MB]

A Visual Validation Method Based on Point Mappings for Medical Image Segmentation
Feng Qian , Lixu Gu, Shaoting Zhang
IEEE 2006 International Conference of the Engineering in Medicine and Biology Society (EMBC2006), New York, USA. Aug., 2006, proceedings
Download: pdf [438 KB]

The Framework for Real-Time Simulation of Deformable Soft-Tissue Using a Hybrid Elastic Model
Shaoting Zhang , Lixu Gu, Weiming Liang, Pengfei Huang , Jan Boehm, and Jianfeng Xu
3rd Symposium on Biomedical Simulation 2006, Zurich, Switzerland. LNCS 4072, Jul., 2006, pp. 75-83.
Download: pdf [1.29 MB]

Real-Time Simulation of Deformable Soft Tissue Based on Mass-Spring and Medial Representation
Shaoting Zhang, Lixu Gu, Pengfei Huang and Jianfeng Xu
Computer Vision for Biomedical Image Applications (a workshop of ICCV2005) , Beijing, P.R.China. LNCS 3765, Oct., 2005, pp. 419-427.
Download: pdf [794 KB] copyrighted

Other Documents

Planning, Training and Guidance for Minimally Invasive Surgery
Plan for my master thesis (12/2006-06/2007).
Download: pdf [360 KB]