- Massively Parallel Approach and Simulation for Medical Image Processing

The recent advances of computer science and technology have increasingly brought innovations in the high performance and efficiency of computing systems. Emerging portable multimedia applications demand more and more tremendous computational throughput with limited energy consumption. Currently, medical imaging applications which are requiring high-efficient and high-throughput embedded processing are becoming an important challenge in computer architecture.

The goal of this project is to propose massively parallel approach and simulation for medical image processing in which new parallel methods are approached to medical imaging analysis using a massively parallel processor array capable of delivering required performance more efficiency. The project proposes parallel approaches for medical image registration, segmentation, and compression using a representative data parallel architecture to accelerate such algorithms. The research contributes a rigorous evaluation of system performance and energy efficiency for the massively parallel processor array against a large set of these medical imaging applications.

Analysis results show that for these algorithms in 130 nm VLSI (SIMD130), a massively parallel processor array can provide 1000x greater computational capabilities and 100x higher energy efficiency than commercial processors including TI DSP C6416 and ARM families (ARM7 and ARM9 series).

References

1. M.Sc. thesis, “Massively Parallel Approach and Simulation for Medical Image Processing”, University of Ulsan, Korea, 2009.
2. Huynh Van Luong, Cheol Hong Kim, and Jong Myon Kim, “A Massively Parallel Algorithm for Fuzzy Vector Quantization,” The KIPS Transactions:PartA, vol. 16-A, no. 6, pp. 411-418, Dec. 2009.
3. Huynh Van Luong and Jong Myon Kim, "A Generalized Spatial Fuzzy C-Means Algorithm for Medical Image Segmentation," in IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2009), Jeju, Korea, Aug. 2009.
4. Huynh Van Luong and Jong Myon Kim, "A Massively Parallel Approach to Affine Transformation in Medical Image Registration," in IEEE International Conference on High Performance Computing and Communications (HPCC-09), Seoul, Korea, Jun. 2009.
5. Huynh Van Luong and Jong Myon Kim, "Parallel Approach to Fuzzy Vector Quantization for Image Compression," in ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD 2009), Daegu, Korea, May 2009.
6. Huynh Van Luong, Sangjin Cho, Jong Myon Kim, and Uipil Chong, "Real-Time Sound Synthesis of Plucked String Instruments using a Data Parallel Architecture," in International Conference on Intelligent Computing (ICIC 2009), Lecture Notes in Computer Science, LNCS 5754, pp. 211-221, 2009.
7. Huynh V. Luong, Choel Hong Kim, and Jong Myon Kim, "Classification of Audio Signal using Generalized Spatial Fuzzy Clustering," in Meeting of Journal of the Acoustical Society of America, pp. 2699-2699, Vol. 125, No. 4, Pt. 2, Portland, Oregon, May 2009.
8. Huynh V. Luong, Sangjin Cho, Jong Myon Kim, and Uipil Chong, "A Real-Time Music Synthesis Engine for Musical Modeling of Plucked String Instruments," in Meeting of Journal of the Acoustical Society of America, pp. 2684-2684, Vol. 125, No. 4, Pt. 2, Portland, Oregon, May 2009.
9. Huynh V. Luong, Jong Myon Kim, and Choel Hong Kim, "Korean articulatory speech synthesis using physical vocal tract model," in Meeting of Journal of the Acoustical Society of America, pp. 2498-2498, Vol. 125, No. 4, Pt. 2, Portland, Oregon, May 2009.
10. Huynh Van Luong and Jong Myon Kim, "A New Parallel Approach to Fuzzy Clustering for Medical Image Segmentation," in International Symposium on Visual Computing (ISVC08), Lecture Notes in Computer Science, LNCS 5358, pp. 1092-1101, 2008.
11. Huynh Van Luong, Jong Myon Kim, and Cheol-Hong Kim, "Performance Optimization of JPEG Software for ARM Processors," in Proceedings of The Engineering and Arts Society in Korea, Daejeon, Korea, Oct. 2008. (Excellent paper award)

Copyright notice: Copyright of the above material belongs to the author. The material may not be copied or re-used without permission from the author.