Texture Analysis Improves Lung-Tissue Segmentation on High-Resolution Computed Tomography in COVID-19

MAZIN Abdalla HASSIB^1,2*Mohammad Elfadil Mohammad Garelnabi³Qurashi Mohammed Ali³Amjad Rashed Alyahyawi²Mamdouh Saud Al-enezi²Mohammed Idris Salih²Ahmed Babikir Abdulla⁴

• ¹Collage of Applied Medical Sciences - Diagnostic radiology Department, University of Hail, Ha'il, Saudi Arabia
• ²University of Hail, Hail, Saudi Arabia
• ³National University - Sudan, Khartoum, Sudan
• ⁴WE care hospital Dammam, Dammam, Saudi Arabia

Background: Accurate separation of lung parenchyma, ground-glass opacity (GGO) and intrapulmonary vessels on high-resolution computed tomography (HRCT) in COVID-19 is challenging. Methods: We conducted a cross-sectional study analysed 530 adults (20–40 years) with RT-PCR–confirmed COVID-19. For texture modelling, we sampled 597 regions of interest (ROIs) representing parenchyma, GGO and intrapulmonary vessels. Region-of-interest–labelled HRCT patches representing parenchyma, GGO and vessels were analyzed using first-and second-order texture features computed across square window sizes (5×5 to 20×20 pixels). Feature selection with stepwise linear discriminant analysis yielded a three-class classifier; the primary endpoint was overall classification accuracy, with secondary endpoints including the effect of window size and identification of the most informative features. Results: The 20×20-pixel window produced the highest performance, with an overall accuracy of 88.6%. Five co-occurrence-based features (difference average, inverse difference moment, co-occurrence matrix standard deviation, sum entropy and information correlation measure 1) were most discriminative; most errors occurred at tissue boundaries where patches spanned mixed voxels. Conclusion: Texture-based feature extraction achieved 88.6% ROI-level accuracy and can serve as a supplementary tool during radiological interpretation of chest CT.