利用田口方法應用於加馬攝影與電腦斷層之影像品質最佳化：V型假體的可行性評估

The Optimization of the SPECT and dual source CT Image Quality via Taguchi Analysis: A feasibility study of V-shaped phantom

沙木瑞(SAMRIT KITTIPAYAK)

指導教授 ： 林政勳

中臺科技大學/健康科學院/醫學影像暨放射科學系暨研究所/博士(2017年)

摘要

Purpose: This work optimized the imaging quality of SPECT and CT from finite analytical data according to Taguchi analysis and to identify significant factors by the analysis of variances (ANOVA).
Materials and methods: The Siemens’s E-cam SPECT system and Siemens SOMATOM Definition Flash CT dual sources with customized V-shaped phantom were used for acquiring practical images to elaborate the analysis. A total of eighteen combinations (L18 orthogonal arrays) of five and six factors were arranged according to Taguchi’s recommendation. Five factors of SPECT system were Scan mode, Collimator type, Matrix size, Zoom and Frame time. Six factors of dual source CT were Kernel, mAs, kVp, Pitch, Slice thickness and Field of View (FOV). The S/N () ratio was specifically defined to satisfy the requirements of optimized spatial resolution of the SPECT and dual source image acquisition in various combinations. Moreover, a statistical ANOVA and F-test were adopted to determine the significant factors.
Results: The optimized values for SPECT image protocol setting were suggested as continuous scan mode, Low Energy High Resolution (LEHR) collimator, 256x256 matrix size, zoom set as 1.45, and frame time for 30 sec. For dual source CT image protocol optimized setting were B40f kernel, 210 mAs, 140 kVp, 1 pitch, slice thickness of 7 mm and FOV 250 mm. Moreover, for SPECT image protocol setting, the dominant factors was only matrix size, the adjust factors were scan mode, collimator type, zoom and frame time. The optimized spatial resolution was upgraded from the original 17.71 mm to 14.23 mm. For dual source CT image protocol optimized setting, all factors were dominant factors, although the verifying dose outputs were slightly larger than those obtained from clinical original setting.
Conclusion: The spatial resolution of the SPECT and dual source CT image were successfully evaluated from the customized V-shaped phantom using the Taguchi analysis. The effectiveness from the L18 orthogonal array helped significantly in reducing the multiple measurements in practical survey.

關鍵字

V-shaped phantom ； Taguchi analysis ； spatial resolution ； dual source CT ； SPECT

並列摘要

Purpose: This work optimized the imaging quality of SPECT and CT from finite analytical data according to Taguchi analysis and to identify significant factors by the analysis of variances (ANOVA).
Materials and methods: The Siemens’s E-cam SPECT system and Siemens SOMATOM Definition Flash CT dual sources with customized V-shaped phantom were used for acquiring practical images to elaborate the analysis. A total of eighteen combinations (L18 orthogonal arrays) of five and six factors were arranged according to Taguchi’s recommendation. Five factors of SPECT system were Scan mode, Collimator type, Matrix size, Zoom and Frame time. Six factors of dual source CT were Kernel, mAs, kVp, Pitch, Slice thickness and Field of View (FOV). The S/N () ratio was specifically defined to satisfy the requirements of optimized spatial resolution of the SPECT and dual source image acquisition in various combinations. Moreover, a statistical ANOVA and F-test were adopted to determine the significant factors.
Results: The optimized values for SPECT image protocol setting were suggested as continuous scan mode, Low Energy High Resolution (LEHR) collimator, 256x256 matrix size, zoom set as 1.45, and frame time for 30 sec. For dual source CT image protocol optimized setting were B40f kernel, 210 mAs, 140 kVp, 1 pitch, slice thickness of 7 mm and FOV 250 mm. Moreover, for SPECT image protocol setting, the dominant factors was only matrix size, the adjust factors were scan mode, collimator type, zoom and frame time. The optimized spatial resolution was upgraded from the original 17.71 mm to 14.23 mm. For dual source CT image protocol optimized setting, all factors were dominant factors, although the verifying dose outputs were slightly larger than those obtained from clinical original setting.
Conclusion: The spatial resolution of the SPECT and dual source CT image were successfully evaluated from the customized V-shaped phantom using the Taguchi analysis. The effectiveness from the L18 orthogonal array helped significantly in reducing the multiple measurements in practical survey.

並列關鍵字

V-shaped phantom ； Taguchi analysis ； spatial resolution ； dual source CT ； SPECT

參考文獻

REFERENCES

1. M.S. George, H.A. Ring, D.C. Costa, et al., Neuroactivation and Neuroimaging with SPECT. Springer-Veriag London, 1991.

2. C.B. Hruska, A.L. Weinmann, M.K. O’Connor, Proof of concept for low-dose molecular breast imaging with a dual-head CZT gamma camera. Part I. evaluation in phantoms. Med Physics 2012; 39: 3466-75.

3. B.L. Holman. Radionuclide Imaging of the Brain. Churchill Livingstone Publishing Co., 1985.

4. R.A. Powsner, E.R. Powsner. Essentials of Nuclear Medicine Physics. Malden, MA: Blackwell Science, 1998.