ABSTRACT
OBJECTIVES: To develop and validate an open-source deep learning model for automatically quantifying scapular and glenoid morphology using CT images of normal subjects and patients with glenohumeral osteoarthritis. MATERIALS AND METHODS: First, we used deep learning to segment the scapula from CT images and then to identify the location of 13 landmarks on the scapula, 9 of them to establish a coordinate system unaffected by osteoarthritis-related changes, and the remaining 4 landmarks on the glenoid cavity to determine the glenoid size and orientation in this scapular coordinate system. The glenoid version, glenoid inclination, critical shoulder angle, glenopolar angle, glenoid height, and glenoid width were subsequently measured in this coordinate system. A 5-fold cross-validation was performed to evaluate the performance of this approach on 60 normal/non-osteoarthritic and 56 pathological/osteoarthritic scapulae. RESULTS: The Dice similarity coefficient between manual and automatic scapular segmentations exceeded 0.97 in both normal and pathological cases. The average error in automatic scapular and glenoid landmark positioning ranged between 1 and 2.5 mm and was comparable between the automatic method and human raters. The automatic method provided acceptable estimates of glenoid version (R2 = 0.95), glenoid inclination (R2 = 0.93), critical shoulder angle (R2 = 0.95), glenopolar angle (R2 = 0.90), glenoid height (R2 = 0.88) and width (R2 = 0.94). However, a significant difference was found for glenoid inclination between manual and automatic measurements (p < 0.001). CONCLUSIONS: This open-source deep learning model enables the automatic quantification of scapular and glenoid morphology from CT scans of patients with glenohumeral osteoarthritis, with sufficient accuracy for clinical use.
Subject(s)
Deep Learning , Osteoarthritis , Scapula , Shoulder Joint , Tomography, X-Ray Computed , Humans , Scapula/diagnostic imaging , Tomography, X-Ray Computed/methods , Osteoarthritis/diagnostic imaging , Male , Female , Shoulder Joint/diagnostic imaging , Middle Aged , Aged , Glenoid Cavity/diagnostic imaging , Adult , Reproducibility of Results , Anatomic Landmarks/diagnostic imaging , Radiographic Image Interpretation, Computer-Assisted/methodsABSTRACT
Background: In reverse shoulder arthroplasty (RSA), the ideal combination of baseplate lateralization (BL), glenosphere size (GS), and glenosphere overhang (GOH) with a commonly used 145° neck shaft angle (NSA) is unclear. This is the first study evaluating correlations of body height (BH), humeral head size (HS), glenoid height (GH), and association of gender with best glenoid configurations for range of motion (ROM) maintaining anatomic lateralization (aLAT) for optimized muscle length in 145° and less distalized 135° RSA. Methods: In this computer model study, 22 computed tomographies without joint narrowing were analyzed (11 male/female). A standardized semi-inlay 145° platform stem was combined with 20 glenoid configurations (baseplate [B] 25, 25 + 3/+6 lateralized [l], 29, 29 + 3/6l combined with glenosphere 36, 36 + 2 eccentric [e], 36 + 3l, 39, 39 + 3e, 39 + 3l , 42, 42 + 4e). Abduction-adduction, flexion-extension, external rotation-internal rotation, total ROM (TROM), and total notching relevant (TNR) ROM were computed, best TROM models respecting aLAT (-1 mm to +1 mm) and HS/GH recorded. Second, the 145° models (Ascend Flex stem; Stryker, Kalamazoo, MI, USA) were converted and compared to a 135° inlay RSA (New Perform stem; Stryker, Kalamazoo, MI, USA) maintaining GOH (6.5-7 mm) and aLAT. Results: Best 145° models had eccentric glenospheres (mean BL: 3.5 mm, GOH 8.8 mm, GS 38.1 mm, distalization 23 mm). The 135° models had concentric glenospheres, mean BL 3.8 mm, GOH 6.9 mm, GS 39.7 mm, and distalization 14.1 mm. HS showed the strongest positive correlation with BL in 145° and 135° models (0.65/0.79). Despite reduced GOH in smaller females with a 135° NSA, adduction, external rotation, extension, TNR ROM, and TROM were significantly increased (P = .02, P = .005, P = .005, P = .004, P = .003), abduction however reduced (P = .02). The same trends were seen for males. Conclusion: HS is a practical measure in surgery or preoperatively, and the strong positive correlation with BL is a useful planning aid. Despite reduction of GOH, conversion to a less distalized 135° NSAinlay design is powerful to maintain and even significantly increase all components of TNR ROM (extension/external rotation/adduction) in small females with the drawback of reduced abduction which may however be compensated by scapula motion. Lateralization with a less distalized 135° RSA optimizes muscle length, may facilitate subscapularis repair, and maintains highest rigid body motion.