Semi-Supervised Deep Learning Semantic Segmentation for 3D Volumetric Computed Tomographic Scoring of Chronic Rhinosinusitis: Clinical Correlations and Comparison with Lund-Mackay Scoring

• Published in: Tomography (Ann Arbor) Vol. 8; no. 2; pp. 718 - 729
• Main Authors: Kuo, Chung-Feng Jeffrey, Liao, Yu-Shu, Barman, Jagadish, Liu, Shao-Cheng
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI 07.03.2022; MDPI AG
• Subjects: artificial intelligence; Deep Learning; Humans; Lund-Mackay score; Maxillary Sinus - diagnostic imaging; MobileNet; Rhinitis - diagnostic imaging; Rhinitis - surgery; Semantics; semi-supervised deep learning; SENet; Supervised Machine Learning; three-dimensional CT; Tomography, X-Ray Computed; Lund-Mackay score; three-dimensional CT; MobileNet; semi-supervised deep learning; SENet; artificial intelligence; ResNet
• ISSN: 2379-139X; 2379-1381; 2379-139X
• DOI: 10.3390/tomography8020059

Background: The traditional Lund-Mackay score (TLMs) is unable to subgrade the volume of inflammatory disease. We aimed to propose an effective modification and calculated the volume-based modified LM score (VMLMs), which should correlate more strongly with clinical symptoms than the TLMs. Methods: Semi-supervised learning with pseudo-labels used for self-training was adopted to train our convolutional neural networks, with the algorithm including a combination of MobileNet, SENet, and ResNet. A total of 175 CT sets, with 50 participants that would undergo sinus surgery, were recruited. The Sinonasal Outcomes Test-22 (SNOT-22) was used to assess disease-specific symptoms before and after surgery. A 3D-projected view was created and VMLMs were calculated for further comparison. Results: Our methods showed a significant improvement both in sinus classification and segmentation as compared to state-of-the-art networks, with an average Dice coefficient of 91.57%, an MioU of 89.43%, and a pixel accuracy of 99.75%. The sinus volume exhibited sex dimorphism. There was a significant positive correlation between volume and height, but a trend toward a negative correlation between maxillary sinus and age. Subjects who underwent surgery had significantly greater TLMs (14.9 vs. 7.38) and VMLMs (11.65 vs. 4.34) than those who did not. ROC-AUC analyses showed that the VMLMs had excellent discrimination at classifying a high probability of postoperative improvement with SNOT-22 reduction. Conclusions: Our method is suitable for obtaining detailed information, excellent sinus boundary prediction, and differentiating the target from its surrounding structure. These findings demonstrate the promise of CT-based volumetric analysis of sinus mucosal inflammation.