Namkug Kim PhD, Presenter: Nothing to Disclose

Joon Beom Seo MD, PhD, Abstract Co-Author: Speaker, Siemens AG

Jeongjin Lee, Abstract Co-Author: Nothing to Disclose

Sujin Park, Abstract Co-Author: Nothing to Disclose

Taekjin Jang, Abstract Co-Author: Nothing to Disclose

Youn Joo Lee MS, Abstract Co-Author: Nothing to Disclose

To compare accuracy and robustness of a new integral-based half-band fitting (IHBF) method with them of a conventional full-width half maximum method (FWHM) in measuring airway wall thickness on CT by using actual and simulated phantoms.

Images were magnified ten-times by bi-cubic interpolation method. For IHBF method, a boundary of thresholded lumen was used as an initial band. From this band, the outward band layers were generated and the mean band density profile was fitted by a Weibull curve to determine the peak robustly. Based on the integral of density inside the peak, half-wall thickness was calculated. Physical phantom, consists of 11 poly-acryl tubes with various inner lumen diameters and thickness, were used for validation. The measured densities of the wall and airspace filled with poly-urethane foam were 150HU and -900HU, respectively. The phantom was scanned in 0.75mm collimation with a 16-multi detector row CT scanner (Siemens Sensation 16). CT images were reconstructed using various reconstruction kernels (B10f, B30f, B50f, B70f and B80f). To mimic complex anatomy of bronchi in human, simulated artificial phantoms of attached vessel and incomplete airway wall were generated and tested. Measured wall thickness was compared with actual thickness and result of convention FWHM method using Bland Altman and paired t-test.

Wall thickness on phantom CT images with soft and standard kernels (B10f, B30f, B50f) using the IHBF method showed better agreement with actual thickness than that of FWHM method (Bland Altman 95% CI by IHBF method: -0.17~-0.01, -0.12~0.14, -0.32~0.18mm vs. by FWHM method:-1.03~0.32, -0.67~0.37, -0.28~0.44mm, respectively). Using the IHBF method, the accuracy was especially higher in measuring wall thinner than 1.2mm (all p<0.05, paired t-test). However, on sharp kernels (B70f, B80f) images, the measurement errors were significantly larger (all p<0.05, paired t-test). In measuring simulated airway phantoms, the IHBF method was more accurate (all p<0.001, paired t-test).

The IHBF method showed more accurate measurement of airway wall thickness in actual and simulated phantom studies than the FWHM method.

This method is especially useful in measuring small airway with wall thinner than 1.2mm and airways with incomplete wall by partial volume artifact or attached vessels than FWHM method.

Kim, N, Seo, J, Lee, J, Park, S, Jang, T, Lee, Y, Integral-based Half-Band Fitting Method in Measuring Airway Wall Thickness on CT: Comparison with Conventional Full-Width Half Maximum Method Using Actual and Simulated Phantoms.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034406.html