Wei Xu

Advanced Applications Engineer, Computational Science Center
Brookhaven National Lab, Upton, NY, 11973

Email: xuw AT bnl DOT gov

I graduated from Computer Science Department of Stony Brook University in Dec. 2012. My advisor was Prof. Klaus Mueller. Before 2013, I worked on Medical Imaging especially GPU-accelerated reconstruction, regularization and restoration methods for Low-dose CT.

Now I am working on research and development of a data processing workflow system for NSLS-II with my supervisor Prof. Dantong Yu.

Selected Research Projects

Stony Brook University
  • Image Database Assisted Low-dose CT Image Restoration
  • (i) Proposed a non-local means based de-noising method employing artifact-free and artifact-matched reference image pairs as priors to overcome the noise and streak artifacts in low-dose CT scans; (ii) devised and implemented image database query framework including SIFT-flow registration, dense SIFT feature generation, kd-tree, k-means clustering, vector quantization and histogram intersection.

  • GPU-accelerated Regularization and Parameter Tuning for Iterative CT Reconstruction
  • (i) Proposed and compared four low-level de-noising filters - bilateral, trilateral, non-local means and adaptive non-local means in the use of reconstruction process in CT. They were applied as non-iterative regularization step replacing traditional total variation minimization; (ii) proposed two parameter learning strategies: exhaustive benchmark test and multi-objective optimization using genetic algorithms. A new perceptual image quality metric was also presented.

  • Iterative Ordered Subset Reconstruction Methods and its Application on GPUs
  • (i) Generalized two algebraic CT reconstruction algorithms into an ordered subset scheme and mapped it to the architecture and programming model of GPUs; (ii) proposed and compared two subset grouping strategies and two relaxation factor selection schemes; (iii) devised a specific framework for Electron Tomography with long object compensation and some performance speedup techniques.

Zhejiang University
  • Image Based Modeling System
  • (i) Devised and implemented several texture mapping algorithms for a 3D image based modeling system based on the geometric relationships between the viewing angle and the surface normal; (ii) implemented two background subtraction algorithms for a marker-free 3D image based modeling system using multiple synchronized cameras.

Selected Publications (full list)

2013
  • new (Fully3D) A Comparative Study of Neighborhood Filters for Artifact Reduction in Iterative Low-Dose CT pdf
  • Wei Xu, Sungsoo Ha, Ziyi Zheng and Klaus Mueller, 12th Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, Lake Tahoe, CA, June 2013

  • new (MedPhy) Database-Assisted Low-dose CT Image Restoration pdf
  • Wei Xu, Sungsoo Ha and Klaus Mueller, Medical Physics, 40(3), 031109, pp. 1-7, 2013

2012

  • (MedPhy) Efficient Low-Dose CT Artifact Mitigation Using An Artifact-Matched Prior Scan pdf
  • Wei Xu and Klaus Mueller, Medical Physics, vol. 39 (8), pp. 4748-4760, 2012

  • (CTMeeting) Database-Assisted Low-dose CT Image Restoration pdf
  • Wei Xu, Sungsoo Ha and Klaus Mueller, The Second International Conference on Image Formation in X-Ray Computed Tomography, Salt Lake City, Utah, June, 2012 (Oral)

2011

  • (Fully3D) A Reference Image Database Approach for NLM Filter-Regularized CT Reconstruction pdf
  • Wei Xu and Klaus Mueller, 11th Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, Potsdam, Germany, pp.116-119, July 2011 (Oral)

  • (GPU GEMS Book Chapter) Using GPUs to Learn Effective Parameter Settings for GPU-Accelerated Iterative CT Reconstruction Algorithms pdf
  • Wei Xu and Klaus Mueller, GPU Computing Gems Emerald Edition, Chapter 43, January 26th 2011

2010

  • (JSB) High-Performance Iterative Electron Tomography Reconstruction with Long-Object Compensation using Graphics Processing Units (GPUs) pdf
  • Wei Xu, Fang Xu, Mel Jones, Bettina Keszthelyi, John Sedat, David Agard and Klaus Mueller, Journal of Structural Biology 171(2):142-153, 2010

  • (CMPB) On the Efficiency of Iterative Ordered Subset Reconstruction Algorithms for Acceleration on GPUs pdf
  • Fang Xu, Wei Xu, Mel Jones, Bettina Keszthelyi, John Sedat, David Agard and Klaus Mueller, Computer Methods and Programs in Biomedicine 98(3):261-270, 2010

  • (MIC) Evaluating Popular Non-Linear Image Processing Filters for their Use in Regularized Iterative CT pdf
  • Wei Xu and Klaus Mueller, IEEE Medical Imaging Conference, Knoxville, Tennessee, October 2010

2009

  • (Fully3D) Learning Effective Parameter Settings for Iterative CT Reconstruction Algorithms pdf

    Wei Xu and Klaus Mueller, 10th Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, pp.251-254, Beijing, China, September 2009 (Oral)

  • (HPIR) A Performance-Driven Study of Regularization Methods for GPU-Accelerated Iterative CT pdf

Wei Xu and Klaus Mueller, 10th Fully3D associated 2nd High Performance Image Reconstruction Workshop, pp.20-23, Beijing, China, September 2009 (Oral, Best paper award)

  • (ISBI) Accelerating Regularized Iterative CT Reconstruction on Commodity Graphics Hardware (GPU) pdf

Wei Xu and Klaus Mueller, IEEE International Symposium on Biomedical Imaging, pp.1287-1290, Boston, MA, June 2009 (Invited, Oral)

Selected Talks

 

Last Updated: 02/14/2013

Hit Counter
Hit Counter
Since March 8, 2013