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

Abstract

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.

Keywords: Lund-Mackay score; MobileNet; ResNet; SENet; artificial intelligence; semi-supervised deep learning; three-dimensional CT.

Figure 1

Segmentation results from comparison between our proposed method and other state-of-the-art networks. U-net misjudged part of the ethmoid sinus as the maxillary sinus. PSPnet had the least fine detailed information. Deeplab-V3+ showed inadequate accuracy in sinus boundary interpretation.

Figure 2

Examples of segmentation and 3D reconstruction of the (A) frontal, (B) maxillary, (C) anterior and posterior ethmoid, and (D) sphenoid sinus. R/L: right/left side, AE/PE: anterior/posterior ethmoid sinus.

Figure 3

(A). A significant positive correlation was found between sinus volume and body height (p < 0.001). The volume had no significant correlations with BMI (B) (p = 0.067) and weight, except for the ethmoid sinus (C) (p = 0.005). (D) A trend toward negative relation between maxillary sinus volume and age (p = 0.053).

Figure 4

Examples of segmentation (A) and 3D reconstruction in CRS patients with air (inner solid part) and opacification (outer hallow part). (B) Frontal, (C) maxillary, (D) anterior and posterior ethmoid, and (E) sphenoid sinus. R/L: right/left side, AE/PE: anterior/posterior ethmoid sinus.

Figure 5

Scatter plots of (A) TLMs and (B) VMLMs values, grouped by whether. surgery was performed or not. ROC with AUC analysis for assessing the correlation between SNOT-22 improvement and the (C) TLMs and (D) VMLMs scoring system.