Computed Tomography Super-Resolution Using a Generative Adversarial Network in Bronchoscopy: A Clinical Feasibility Study

To evaluate the usefulness of applying computed tomography (CT) images reconstructed by a deep learning super-resolution method to the clinical scenario of planning a real bronchoscopy procedure. We trained a super-resolution generative adversarial network (SRGAN) to reconstruct CT images to high-resolution (SRGANrc). We tasked three pulmonologists with evaluating the quality of the CT images and the derived virtual bronchoscopies. We also compared the number of bronchi that were segmented by an automatic commercial program with the number of bronchi segmented in different processed thin-sectioned CT images. Regarding the human visual score, the original thin-sectioned CT images received more votes than the reconstructed CT images (SRGANrc) (29 votes versus eight votes). As for the human classification of four high-resolution CT images, the majority of images (83.7%) were classified correctly. Four out of 23 virtual bronchoscopies derived from super-resolution CT images were considered superior. The number of automatically segmented bronchi in super-resolution CT images was on average 1.5 less than that in the original thin-sliced CT images (mean bronchi: 15.1 vs. 16.6). The reconstruction of super-resolution CT images through the SRGAN may have limited applications in the clinical scenarios of our study. In addition to improving the deep-learning algorithm, we need more clinical implementation tests to discover its value.

[1]  Liying Zheng,et al.  Medical Image Super Resolution Using Improved Generative Adversarial Networks , 2019, IEEE Access.

[2]  J. Kurhanewicz,et al.  High-Resolution 3-T Endorectal Prostate MRI: A Multireader Study of Radiologist Preference and Perceived Interpretive Quality of 2D and 3D T2-Weighted Fast Spin-Echo MR Images. , 2016, AJR. American journal of roentgenology.

[3]  Pietro Lio',et al.  How Can We Make Gan Perform Better in Single Medical Image Super-Resolution? A Lesion Focused Multi-Scale Approach , 2019, 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).

[4]  Christian Ledig,et al.  Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Naoki Yamamoto,et al.  Bronchus sign on thin-section computed tomography is a powerful predictive factor for successful transbronchial biopsy using endobronchial ultrasound with a guide sheath for small peripheral lung lesions: a retrospective observational study , 2015, BMC Medical Imaging.

[6]  Stephen M. Smith,et al.  A global optimisation method for robust affine registration of brain images , 2001, Medical Image Anal..

[7]  Dwarikanath Mahapatra,et al.  Image super-resolution using progressive generative adversarial networks for medical image analysis , 2019, Comput. Medical Imaging Graph..

[8]  European Society of Radiology Usability of irreversible image compression in radiological imaging. A position paper by the European Society of Radiology (ESR) , 2011, Insights into imaging.

[9]  Shinji Sasada,et al.  Diagnostic utility of endobronchial ultrasound with a guide sheath under the computed tomography workstation (ziostation) for small peripheral pulmonary lesions , 2017, The clinical respiratory journal.

[10]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

[11]  Jithin Saji Isaac,et al.  Super resolution techniques for medical image processing , 2015, 2015 International Conference on Technologies for Sustainable Development (ICTSD).