Combination of Window-Modulated Ultrasound Nakagami Imaging and Gaussian Approximation for Radiofrequency Ablation Monitoring: An In Vitro Study
- 44 Downloads
The ultrasound Nakagami parameter map efficiently describes the statistical properties of the liver tissue. Using sliding windows in Nakagami imaging for monitoring radiofrequency ablation (RFA) is feasible. However, a limitation of conventional Nakagami imaging is the inability to visualize precise ablation (necrosis) zones by using a single sliding window. To accurately estimate ablation zones, we used window-modulated compounding (WMC) Nakagami imaging based on the synthesis of Nakagami images by using windows of different side lengths. Moreover, we proposed a new strategy for the postprocessing of Nakagami images by using Gaussian approximation (GAX) to approximate the cross-sectional areas of ablation zones. We heated 15 porcine liver samples for 12 min by using electrodes of tip lengths 0.5, 1.0, and 1.5 cm to ensure tissue necrosis after the ablation. The contour area of −6 dB in two-dimensional (2D) GAX images increased from 41 to 209 mm2, whereas that in the measures of section images increased from 40 to 265 mm2, with an increase in the electrode tips length from 0.5 to 1.5 cm. The coefficient of determination between the contour area of −6 dB in 2D GAX images and that in the measures of section images was 0.97, indicating a more satisfactory performance of GAX compared with that of polynomial approximation. The area estimated using GAX was similar to a circle, which conformed to the physical process of RFA.
KeywordsUltrasound Compounding Nakagami imaging Radiofrequency ablation Gaussian approximation
The use of facilities was supported by Ultrasound Imaging Lab, Department of Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan. This work was also supported by the National Natural Science Foundation of China (No. 61471263).
- 7.Sokka, S., King, R., McDannold, N., & Hynynen, K. (2002). Gas-bubble enhanced heating in rabbit thigh in vivo. IEEE Ultrasonics Symposium, 2, 1415–1418.Google Scholar
- 9.Szabo, T. L. (2004). Diagnostic ultrasound imaging. Burlington, MA: Elsevier Academic Press.Google Scholar
- 10.Shung, K. K., & Smith, M. B. (1992). Principles of medical imaging. San Diego: Academic Press.Google Scholar
- 23.Hoffman, H. C. (1960). Statistical methods on radio wave propagation. New York: Pergamon Press.Google Scholar
- 27.J. B. Hampshire, II. (1988). A non-Rayleigh model for ultrasonic backscatter in myocardium, Hanover NH: M. S. thesis, Thayer School Eng., Darmouth College.Google Scholar
- 28.J. B. Hampshire, J. W. Strohbehn, M. D. McDaniel, J. L. Waugh and D. H. James. (1988). Probability density of myocardial ultrasonic backscatter. In: IEEE Proceedings of the 14th Annual Northeast Bioengineering Conference (pp. 305–308).Google Scholar
- 35.Larrue, A. & Noble, J. A. (2011). Nakagami imaging with small windows. In: IEEE Ultrasonic Symposium Proceedings (pp. 887–890).Google Scholar
- 36.Degroot, M. H., & Schervish, M. J. (2011). Probability and statistics (4th ed.). San Antonio: Pearson Education Inc.Google Scholar
- 44.Tsui, P. H., Wan, Y. L., Tai, D. I., & Shu, Y. C. (2015). Effects of estimators on ultrasound Nakagami imaging in visualizing the change in the backscattered statistics from a Rayleigh distribution to a pre-Rayleigh distribution. Ultrasound in Medicine and Biology, 41, 2240–2251.CrossRefGoogle Scholar