Determination of pre-arthropathy scapular anatomy with a statistical shape model: part I-rotator cuff tear arthropathy

A scapular statistical shape model can reliably predict premorbid glenoid morphology in conditions of severe glenoid bone loss

BACKGROUND

Knowledge of premorbid glenoid parameters at the time of shoulder arthroplasty, such as inclination, version, joint line position, height, and width, can assist with implant selection, implant positioning, metal augment sizing, and/or bone graft dimensions. The objective of this study was to validate a scapular statistical shape model (SSM) in predicting patient-specific glenoid morphology in scapulae with clinically relevant glenoid erosion patterns.

METHODS

Computed tomography scans of 30 healthy scapulae were obtained and used as the control group. Each scapula was then virtually eroded to create 7 erosion patterns (Walch A1, A2, B2, B3, D, Favard E2, and E3). This resulted in 210 uniquely eroded glenoid models, forming the eroded glenoid group. A scapular SSM, created from a different database of 85 healthy scapulae, was then applied to each eroded scapula to predict the premorbid glenoid morphology. The premorbid glenoid inclination, version, height, width, radius of best-fit sphere, and glenoid joint line position were automatically calculated for each of the 210 eroded glenoids. The mean values for all outcome variables were compared across all erosion types between the healthy, eroded, and SSM-predicted groups using a 2-way repeated measures analysis of variance.

RESULTS

The SSM was able to predict the mean premorbid glenoid parameters of the eroded glenoids with a mean absolute difference of 3° ± 2° for inclination, 3° ± 2° for version, 2 ± 1 mm for glenoid height, 2 ± 1 mm for glenoid width, 5 ± 4 mm for radius of best-fit sphere, and 1 ± 1 mm for glenoid joint line. The mean SSM-predicted values for inclination, version, height, width, and radius were not significantly different than the control group (P > .05).

DISCUSSION

An SSM has been developed that can reliably predict premorbid glenoid morphology and glenoid indices in patients with common glenoid erosion patterns. This technology can serve as a useful template to visually represent the premorbid healthy glenoid in patients with severe glenoid bony erosions. Knowledge of the premorbid glenoid preoperatively can assist with implant selection, positioning, and sizing.

Covariations between scapular shape and bone density in B-glenoids: A statistical shape and density modeling-approach

B-type glenoids are characterized by posterior humeral head migration and/or bony-erosion-induced glenoid retroversion. Patients with this type of osteoarthritic glenoids are known to be at increased risk of glenoid component loosening after anatomic total shoulder arthroplasty (aTSA). One of the main challenges in B glenoid surgical planning is to find a balance between correcting the bony shape and maintaining the quality of the bone support. This study aims to systematically quantify variabilities in terms of scapular morphology and bone mineral density in patients with B glenoids and to identify patterns of covariation between these two features. Using computed tomography scan images of 62 patients, three-dimensional scapular surface models were constructed. Rigid and nonrigid surface registration of the scapular surfaces, followed by volumetric registration and material mapping, enabled us to develop statistical shape model (SSM) and statistical density model (SDM). Partial least square correlation (PLSC) was used to identify patterns of covariation. The developed SSM and SDM represented 85.9% and 56.6% of variabilities in terms of scapular morphology and bone density, respectively. PLSC identified four modes of covariation, explaining 66.0% of the correlation between these two variations. Covariation of posterior-inferior glenoid erosion with posterior sclerotic bone formation in association with reduction of bone density in the anterior and central part of the glenoid was detected as the primary mode of covariation. Identification of these asymmetrical distribution of bone density can inform us about possible reasons behind glenoid component loosening in B glenoids and surgical guidelines in terms of the compromise between bony shape correction and bone support quality.

Quantitative statistical shape model-based analysis of humeral head migration, Part 2: Shoulder osteoarthritis

We wanted to investigate the quantitative characteristics of humeral head migration (HHM) in shoulder osteoarthritis (OA) and their possible associations with scapular morphology. We quantified CT-scan-based-HHM in 122 patients with a combination of automated 3D scapulohumeral migration (=HHM with respect to the scapula) and glenohumeral migration (=HHM with respect to the glenoid) measurements. We divided OA patients in Group 1 (without HHM), Group 2a (anterior HHM) and Group 2b (posterior HHM). We reconstructed and measured the prearthropathy scapular anatomy with a statistical shape model technique. HHM primarily occurs in the axial plane in shoulder OA. We found "not-perfect" correlation between subluxation distance AP and scapulohumeral migration values (r_s  = 0.8, p < 0.001). Group 2b patients had a more expressed prearthropathy glenoid retroversion (13° vs. 7°, p < 0.001) and posterior glenoid translation (4 mm vs. 6 mm, p = 0.003) in comparison to Group 1. Binary logistic regression analysis indicated prearthropathy glenoid version as a significant predictor of HHM (χ² = 27, p < 0.001). Multivariate regression analysis showed that the pathologic version could explain 56% of subluxation distance-AP variance and 75% of the scapulohumeral migration variance (all p < 0.001). Herewith, every degree increase in pathologic glenoid retroversion was associated with an increase of 1% subluxation distance-AP, and scapulohumeral migration. The occurrence of posterior HHM is associated with prearthropathy glenoid retroversion and more posterior glenoid translation. The reported regression values of HHM in the function of the pathologic glenoid version could form a basis toward a more patient-specific correction of HHM.

A predictive model for the critical shoulder angle based on a three-dimensional analysis of scapular angular and linear morphometrics

BACKGROUND

The purpose of this study was to define the features of scapular morphology that are associated with changes in the critical shoulder angle (CSA) by developing the best predictive model for the CSA based on multiple potential explanatory variables, using a completely 3D assessment.

METHODS

3D meshes were created from CT DICOMs using InVesalius (Vers 3.1.1, RTI [Renato Archer Information Technology Centre], Brazil) and Meshmixer (3.4.35, Autodesk Inc., San Rafael, CA). The analysis included 17 potential angular, weighted linear and area measurements. The correlation of the explanatory variables with the CSA was investigated with the Pearson's correlation coefficient. Using multivariable linear regression, the approach for predictive model-building was leave-one-out cross-validation and best subset selection.

RESULTS

Fifty-three meshes were analysed. Glenoid inclination (GI) and coronal plane angulation of the acromion (CPAA) [Pearson's r: 0.535; -0.502] correlated best with CSA. The best model (adjusted R-squared value 0.67) for CSA prediction contained 10 explanatory variables including glenoid, scapular spine and acromial factors. CPAA and GI were the most important based on their distribution, estimate of coefficients and loss in predictive power if removed.

CONCLUSIONS

The relationship between scapular morphology and CSA is more complex than the concept of it being dictated solely by GI and acromial horizontal offset and includes glenoid, scapular spine and acromial factors of which CPAA and GI are most important. A further investigation in a closely defined cohort with rotator cuff tears is required before drawing any clinical conclusions about the role of surgical modification of scapular morphology.

LEVEL OF EVIDENCE

Level 4 retrospective observational cohort study with no comparison group.

A 3-Dimensional Classification for Degenerative Glenohumeral Arthritis Based on Humeroscapular Alignment

Background

Seminal classifications of degenerative arthritis of the shoulder (DAS) describe either cuff tear arthropathy in the coronal plane or primary osteoarthritis in the cross-sectional plane. None consider a biplanar eccentricity.

Purpose/Hypothesis

The purpose of this study was to investigate humeroscapular alignment (HSA) of patients with DAS in both the anteroposterior (A-P) and superoinferior (S-I) planes on computed tomography (CT) after 3-dimensional (3D) reconstruction and develop a classification based on biplanar HSA in 9 quadrants. It was hypothesized that biplanar eccentricity would occur frequently.

Study Design

Cross-sectional study; Level of evidence, 3.

Methods

The authors analyzed 130 CT scans of patients who had undergone shoulder arthroplasty. The glenoid center, trigonum, and inferior angle of the scapula were aligned in a single plane using 3D reconstruction software. Subluxation of the HSA was measured as the distance from the center of rotation of the humeral head to the scapular axis (line from trigonum through glenoid center) and was expressed as a percentage of the radius of the humeral head in both the A-P and the S-I directions. HSA was described in terms of A-P alignment first (posterior/central/anterior), then S-I alignment (superior/central/inferior), for a total of 9 different alignment combinations. Additionally, glenoid erosion was graded 1-3.

Results

Subluxation of the HSA was 74.1% posterior to 23.5% anterior in the A-P direction and 17.2% inferior to 68.6% superior in the S-I direction. A central HSA was calculated as between 20% posterior to 5% anterior (A-P) and 5% inferior to 20% superior (S-I), after a graphical analysis. Posterior subluxation >60% of the radius was labeled as extraposterior, and static acetabularization was labeled as extrasuperior. Overall, 21 patients had central-central, 40 centrosuperior, and 1 centroinferior alignment. Of 60 shoulders with posterior subluxation, alignment was posterocentral in 31, posterosuperior in 25, and posteroinferior in 5. There were 3 patients with anterocentral and 4 anterosuperior subluxation; in addition, 4 cases with extraposterior and 17 with extrasuperior subluxation were identified.

Conclusion

There was a high prevalence of biplanar eccentricity in DAS. The 3D classification system using combined HSA and glenoid erosion can be applied to describe DAS comprehensively.

Quantitative SSM-based analysis of humeral head migration in rotator cuff tear arthropathy patients

Rotator cuff tear arthropathy (RCTA) is characterized by massive rotator cuff tearing combined with humeral head migration (HHM). The aim of this study is to investigate the quantitative characteristics of this migration and its association with glenoid erosions and prearthropathy scapular anatomy. We quantified HHM and prearthropathy scapular anatomy of 64 RCTA patients with statistical shape modeling-based techniques. Glenoid erosion was classified according to Sirveaux et al. A cutoff value for confirming HHM was 5 mm based on a control group of 49 patients. Group 1 (RCTA without HHM) consisted of 21 patients, with a mean subluxation distance (SLD) of 3 mm. Group 2 (RCTA with HHM) consisted of 43 patients, with mean SLD of 9 mm, SLD in the anteroposterior plane of -1 mm (SD ± 4 mm), SLD in the superoinferior plane of 7 mm (SD ± 3 mm), and subluxation angle (SLA) of -5° (SD ± 40°). Analysis with Fisher's exact test showed a clear association between HHM and glenoid erosions (p = 0.002). Multivariate regression analysis of Group 2 showed that prearthropathy lateral acromial angle combined with critical shoulder angle (p = 0.004) explained 21% of the observed variability in SLD. The prearthropathy glenoid version explained 23% of the variability in SLA (p = 0.001). HHM in RCTA patients has a wide variation in both magnitude and direction leading to a distorted glenohumeral relationship in the coronal and axial plane. HHM is highly associated with the occurrence of glenoid erosions. There is a correlation between the prearthropathy scapular anatomy and the magnitude and direction of HHM.

Exploring palatal and dental shape variation with 3D shape analysis and geometric deep learning

OBJECTIVES

Palatal shape contains a lot of information that is of clinical interest. Moreover, palatal shape analysis can be used to guide or evaluate orthodontic treatments. A statistical shape model (SSM) is a tool that, by means of dimensionality reduction, aims at compactly modeling the variance of complex shapes for efficient analysis. In this report, we evaluate several competing approaches to constructing SSMs for the human palate.

SETTING AND SAMPLE POPULATION

This study used a sample comprising digitized 3D maxillary dental casts from 1,324 individuals.

MATERIALS AND METHODS

Principal component analysis (PCA) and autoencoders (AE) are popular approaches to construct SSMs. PCA is a dimension reduction technique that provides a compact description of shapes by uncorrelated variables. AEs are situated in the field of deep learning and provide a non-linear framework for dimension reduction. This work introduces the singular autoencoder (SAE), a hybrid approach that combines the most important properties of PCA and AEs. We assess the performance of the SAE using standard evaluation tools for SSMs, including accuracy, generalization, and specificity.

RESULTS

We found that the SAE obtains equivalent results to PCA and AEs for all evaluation metrics. SAE scores were found to be uncorrelated and provided an optimally compact representation of the shapes.

CONCLUSION

We conclude that the SAE is a promising tool for 3D palatal shape analysis, which effectively combines the power of PCA with the flexibility of deep learning. This opens future AI driven applications of shape analysis in orthodontics and other related clinical disciplines.