Three-dimensional MRI Bone Models of the Glenohumeral Joint Using Deep Learning: Evaluation of Normal Anatomy and Glenoid Bone Loss

From Labral Tears to Bone Loss: Imaging the Spectrum of Shoulder Instability

Glenohumeral instability results from disruptions in dynamic and static stabilizers of the glenohumeral joint, leading to dislocation, subluxation, or chronic apprehension. Although anterior glenohumeral instability is the most common form, posterior instability represents a distinct and often underrecognized condition with unique biomechanical and clinical characteristics. Imaging plays a critical role in evaluating shoulder instability, guiding diagnosis, and informing treatment decisions. Radiographs, computed tomography, and magnetic resonance imaging help assess structural abnormalities, such as labral tears, capsuloligamentous injuries, and osseous defects. Advanced imaging techniques, such as three-dimensional computed tomography and zero echo time magnetic resonance imaging, offer improved assessment of bone loss; magnetic resonance arthrography enhances the detection of soft tissue pathology. This article offers a comprehensive review of the imaging spectrum of glenohumeral instability, covering labral pathology, capsuloligamentous injuries, and bone loss. It highlights the critical role of precise imaging assessment in guiding optimal management strategies for both anterior and posterior instability patterns.

Shoulder Arthroplasty: Preoperative Evaluation and Postoperative Imaging

Shoulder arthroplasty procedures have increased over the past few years. Several different arthroplasty options are available for varying indications, such as humeral head resurfacing, hemiarthroplasty, anatomical total shoulder arthroplasty, and reverse total shoulder arthroplasty, with ongoing modifications of prosthesis components and surgical techniques. Arthroplasty complications are encountered from the acute postoperative period to several years postoperatively. This article reviews the more common types of shoulder arthroplasties: their imaging appearances, multimodality imaging assessments for preoperative planning, and complications.

Deep learning algorithms enable MRI-based scapular morphology analysis with values comparable to CT-based assessments

Scapular morphological attributes show promise as prognostic indicators of retear following rotator cuff repair. Current evaluation techniques using single-slice magnetic-resonance imaging (MRI) are, however, prone to error, while more accurate computed tomography (CT)-based three-dimensional techniques, are limited by cost and radiation exposure. In this study we propose deep learning-based methods that enable automatic scapular morphological analysis from diagnostic MRI despite the anisotropic resolution and reduced field of view, compared to CT. A deep learning-based segmentation network was trained with paired CT derived scapula segmentations. An algorithm to fuse multi-plane segmentations was developed to generated high-resolution 3D models of the scapula on which morphological landmark- and axes were predicted using a second deep learning network for morphological analysis. Using the proposed methods, the critical shoulder angle, glenoid inclination and version were measured from MRI with accuracies of -1.3 ± 1.7 degrees, 1.3 ± 2.1 degree, and - 1.4 ± 3.4 degrees respectively, compared to CT. Inter-class correlation between MRI and CT derived metrics were substantial for the glenoid version and almost perfect for the other metrics. This study demonstrates how deep learning can overcome reduced resolution, bone border contrast and field of view, to enable 3D scapular morphology analysis on MRI.

The diagnostic value of MRI segmentation technique for shoulder joint injuries based on deep learning

This work is to investigate the diagnostic value of a deep learning-based magnetic resonance imaging (MRI) image segmentation (IS) technique for shoulder joint injuries (SJIs) in swimmers. A novel multi-scale feature fusion network (MSFFN) is developed by optimizing and integrating the AlexNet and U-Net algorithms for the segmentation of MRI images of the shoulder joint. The model is evaluated using metrics such as the Dice similarity coefficient (DSC), positive predictive value (PPV), and sensitivity (SE). A cohort of 52 swimmers with SJIs from Guangzhou Hospital serve as the subjects for this study, wherein the accuracy of the developed shoulder joint MRI IS model in diagnosing swimmers' SJIs is analyzed. The results reveal that the DSC for segmenting joint bones in MRI images based on the MSFFN algorithm is 92.65%, with PPV of 95.83% and SE of 96.30%. Similarly, the DSC for segmenting humerus bones in MRI images is 92.93%, with PPV of 95.56% and SE of 92.78%. The MRI IS algorithm exhibits an accuracy of 86.54% in diagnosing types of SJIs in swimmers, surpassing the conventional diagnostic accuracy of 71.15%. The consistency between the diagnostic results of complete tear, superior surface tear, inferior surface tear, and intratendinous tear of SJIs in swimmers and arthroscopic diagnostic results yield a Kappa value of 0.785 and an accuracy of 87.89%. These findings underscore the significant diagnostic value and potential of the MRI IS technique based on the MSFFN algorithm in diagnosing SJIs in swimmers.

The Development of a Yolov8-Based Model for the Measurement of Critical Shoulder Angle (CSA), Lateral Acromion Angle (LAA), and Acromion Index (AI) from Shoulder X-ray Images

Background: The accurate and effective evaluation of parameters such as critical shoulder angle, lateral acromion angle, and acromion index from shoulder X-ray images is crucial for identifying pathological changes and assessing disease risk in the shoulder joint. Methods: In this study, a YOLOv8-based model was developed to automatically measure these three parameters together, contributing to the existing literature. Initially, YOLOv8 was used to segment the acromion, glenoid, and humerus regions, after which the CSA, LAA angles, and AI between these regions were calculated. The MURA dataset was employed in this study. Results: Segmentation performance was evaluated with the Dice and Jaccard similarity indices, both exceeding 0.9. Statistical analyses of the measurement performance, including Pearson correlation coefficient, RMSE, and ICC values demonstrated that the proposed model exhibits high consistency and similarity with manual measurements. Conclusions: The results indicate that automatic measurement methods align with manual measurements with high accuracy and offer an effective alternative for clinical applications. This study provides valuable insights for the early diagnosis and management of shoulder diseases and makes a significant contribution to existing measurement methods.

Reliability of glenoid measurements performed using Multiplanar Reconstruction (MPR) of Magnetic Resonance (MRI) in patients with shoulder instability

PURPOSE

Measurement of glenoid bone loss in the shoulder instability can be assessed by CT or MRI multiplanar imaging and is crucial for pre-operative planning. The aim of this study is to determine the intra and interobserver reliability of glenoid deficiency measurement using MRI multiplanar reconstruction with 2D assessment in the sagittal plane (MPR MRI).

METHODS

We reviewed MRI images of 80 patients with anterior shoulder instability with Osirix software using MPR. Six observers with basic experience measured the glenoid, erosion edge length, and bone loss twice, with at least one-week interval between measurements. We calculated reliability and repeatability using the intra-class correlation coefficient (ICC) and minimal detectable change with 95% confidence (MDC95%).

RESULTS

Intra and Inter-observer ICC and MDC95% for glenoid width and height were excellent (ICC 0,89-0,96). For erosion edge length and area of the glenoid were acceptable/good (ICC 0,61-0,89). Bone loss and Pico Index were associated with acceptable/good ICC (0,63 -0,86)) but poor MDC95% (45 - 84 %). Intra-observer reliability improved with time, while inter-observer remained unchanged.

CONCLUSION

The MPR MRI measurement of the anterior glenoid lesion is very good tool for linear parameters. This method is not valid for Pico index measurement, as the area of bone loss is variable. The pace of learning is individual, therefore complex calculations based on MPR MRI are not resistant to low experience as opposed to true 3D CT.

Artificial intelligence in shoulder and elbow surgery: overview of current and future applications

Artificial intelligence (AI) is amongst the most rapidly growing technologies in orthopedic surgery. With the exponential growth in healthcare data, computing power, and complex predictive algorithms, this technology is poised to aid providers in data processing and clinical decision support throughout the continuum of orthopedic care. Understanding the utility and limitations of this technology is vital to practicing orthopedic surgeons, as these applications will become more common place in everyday practice. AI has already demonstrated its utility in shoulder and elbow surgery for imaging-based diagnosis, predictive modeling of clinical outcomes, implant identification, and automated image segmentation. The future integration of AI and robotic surgery represents the largest potential application of AI in shoulder and elbow surgery with the potential for significant clinical and financial impact. This editorial's purpose is to summarize common AI terms, provide a framework to understand and interpret AI model results, and discuss current applications and future directions within shoulder and elbow surgery.

Shoulder Bone Segmentation with DeepLab and U-Net

Evaluation of 3D bone morphology of the glenohumeral joint is necessary for pre-surgical planning. Zero echo time (ZTE) magnetic resonance imaging (MRI) provides excellent bone contrast and can potentially be used in place of computed tomography. Segmentation of shoulder anatomy, particularly humeral head and acetabulum, is needed for detailed assessment of each anatomy and for pre-surgical preparation. In this study we compared performance of two popular deep learning models based on Google's DeepLab and U-Net to perform automated segmentation on ZTE MRI of human shoulders. Axial ZTE images of normal shoulders (n=31) acquired at 3-Tesla were annotated for training with a DeepLab and 2D U-Net, and the trained model was validated with testing data (n=13). While both models showed visually satisfactory results for segmenting the humeral bone, U-Net slightly over-estimated while DeepLab under-estimated the segmented area compared to the ground truth. Testing accuracy quantified by Dice score was significantly higher (p<0.05) for U-Net (88%) than DeepLab (81%) for the humeral segmentation. We have also implemented the U-Net model onto an MRI console for a push-button DL segmentation processing. Although this is an early work with limitations, our approach has the potential to improve shoulder MR evaluation hindered by manual post-processing and may provide clinical benefit for quickly visualizing bones of the glenohumeral joint.

Application of Artificial Intelligence in Shoulder Pathology

Artificial intelligence (AI) refers to the science and engineering of creating intelligent machines for imitating and expanding human intelligence. Given the ongoing evolution of the multidisciplinary integration trend in modern medicine, numerous studies have investigated the power of AI to address orthopedic-specific problems. One particular area of investigation focuses on shoulder pathology, which is a range of disorders or abnormalities of the shoulder joint, causing pain, inflammation, stiffness, weakness, and reduced range of motion. There has not yet been a comprehensive review of the recent advancements in this field. Therefore, the purpose of this review is to evaluate current AI applications in shoulder pathology. This review mainly summarizes several crucial stages of the clinical practice, including predictive models and prognosis, diagnosis, treatment, and physical therapy. In addition, the challenges and future development of AI technology are also discussed.

Glenoid segmentation from computed tomography scans based on a 2-stage deep learning model for glenoid bone loss evaluation

BACKGROUND

The best-fitting circle drawn by computed tomography (CT) reconstruction of the en face view of the glenoid bone to measure the bone defect is widely used in clinical application. However, there are still some limitations in practical application, which can prevent the achievement of accurate measurements. This study aimed to accurately and automatically segment the glenoid from CT scans based on a 2-stage deep learning model and to quantitatively measure the glenoid bone defect.

MATERIALS AND METHODS

Patients who were referred to our institution between June 2018 and February 2022 were retrospectively reviewed. The dislocation group consisted of 237 patients with a history of ≥2 unilateral shoulder dislocations within 2 years. The control group consisted of 248 individuals with no history of shoulder dislocation, shoulder developmental deformity, or other disease that may lead to abnormal morphology of the glenoid. All patients underwent CT examination with a 1-mm slice thickness and a 1-mm increment, including complete imaging of the bilateral glenoid. A residual neural network (ResNet) location model and a U-Net bone segmentation model were constructed to develop an automated segmentation model for the glenoid from CT scans. The data set was randomly divided into training (201 of 248) and test (47 of 248) data sets of control-group data and training (190 of 237) and test (47 of 237) data sets of dislocation-group data. The accuracy of the stage 1 (glenoid location) model, the mean intersection-over-union value of the stage 2 (glenoid segmentation) model, and the glenoid volume error were used to assess the performance of the model. The R2 value and Lin concordance correlation coefficient were used to assess the correlation between the prediction and the gold standard.

RESULTS

A total of 73,805 images were obtained after the labeling process, and each image was composed of CT images of the glenoid and its corresponding mask. The average overall accuracy of stage 1 was 99.28%; the average mean intersection-over-union value of stage 2 was 0.96. The average glenoid volume error between the predicted and true values was 9.33%. The R2 values of the predicted and true values of glenoid volume and glenoid bone loss (GBL) were 0.87 and 0.91, respectively. The Lin concordance correlation coefficient value of the predicted and true values of glenoid volume and GBL were 0.93 and 0.95, respectively.

CONCLUSION

The 2-stage model in this study showed a good performance in glenoid bone segmentation from CT scans and could quantitatively measure GBL, providing a data reference for subsequent clinical treatment.

Glenohumeral osteoarthritis: what the surgeon needs from the radiologist

Glenohumeral osteoarthritis (GHOA) is a widely prevalent disease with increasing frequency due to population aging. Both clinical manifestations and radiography play key roles in the initial diagnosis, staging, and management decisions. Radiographic disease progression evaluation is performed using validated staging systems, such as Kellgren and Lawrence, Samilson, and Hamada. For young patients with mild to moderate GHOA and failed conservative treatment, arthroscopic preservation surgery (APS) is usually considered. Older patients and those with severe GHOA benefit from different types of arthroplasties. Preoperative magnetic resonance imaging (MRI) is essential for APS surgical planning, as it maps repairable labral, cartilage, and rotator cuff lesions. For arthroplasty planning, the status of glenoid cartilage and intactness of rotator cuff as well as glenoid morphology represent key factors guiding the decision regarding the most suitable hardware design, whether resurfacing, partial, total, or reverse joint replacement. Pre-surgical MRI or alternatively computed tomography arthrogram is employed to evaluate the cartilage and rotator cuff. Finally, three-dimensional computed tomography (3D CT) is indicated to optimally assess the glenoid morphology (to determine Walch classification, version, inclination, and bone loss) and analyze the necessity for glenoid osteotomy or graft augmentation to correct the glenoid structural abnormalities for future success and longevity of the shoulder implants or chosen constructs. Understanding the purpose of each imaging and treatment modality allows more efficient image interpretation. This article reviews the above concepts and details what a surgeon needs from a radiologist and could benefit from accurate reporting of preoperative imaging studies.

Comparing bone shape models from deep learning processing of magnetic resonance imaging to computed tomography-based models

Background

The purpose of this study was to develop a deep learning approach to automatically segment the scapular bone on magnetic resonance imaging (MRI) images and to compare the accuracy of these three-dimensional (3D) models with that of 3D computed tomography (CT).

Methods

Fifty-five patients with high-resolution 3D fat-saturated T2 MRI were retrospectively identified. The underlying pathology included rotator cuff tendinopathy and tears, shoulder instability, and impingement. Two experienced musculoskeletal researchers manually segmented the scapular bone. Five cross-validation training and validation splits were generated to independently train two-dimensional (2D) and 3D models using a convolutional neural network approach. Model performance was evaluated using the Dice similarity coefficient (DSC). All models with DSC > 0.70 were ensembled and used for the test set, which consisted of four patients with matching high-resolution MRI and CT scans. Clinically relevant glenoid measurements, including glenoid height, width, and retroversion, were calculated for two of the patients. Paired t-tests and Wilcoxon signed-rank tests were used to compare the DSC of the models.

Results

The 2D and 3D models achieved a best DSC of 0.86 and 0.82, respectively, with no significant difference observed. Augmentation of imaging data significantly improved 3D but not 2D model performance. In comparing clinical measurements of 3D MRI and CT, there was a mean difference ranging from 1.29 mm to 3.46 mm and 0.05° to 7.47°.

Conclusion

We have presented a fully automatic, deep learning-based strategy for extracting scapular shape from a high-resolution MRI scan. Further developments of this technology have the potential to allow for surgeons to obtain all clinically relevant information from MRI scans and reduce the need for multiple imaging studies for patients with shoulder pathology.

Artificial intelligence-based applications in shoulder surgery leaves much to be desired: a systematic review

Background

Artificial intelligence (AI) aims to simulate human intelligence using automated computer algorithms. There has been a rapid increase in research applying AI to various subspecialties of orthopedic surgery, including shoulder surgery. The purpose of this review is to assess the scope and validity of current clinical AI applications in shoulder surgery literature.

Methods

A systematic literature review was conducted using PubMed for all articles published between January 1, 2010 and June 10, 2022. The search query used the terms as follows: (artificial intelligence OR machine learning OR deep learning) AND (shoulder OR shoulder surgery OR rotator cuff). All studies that examined AI application models in shoulder surgery were included and evaluated for model performance and validation (internal, external, or both).

Results

A total of 45 studies were included in the final analysis. Eighteen studies involved shoulder arthroplasty, 13 rotator cuff, and 14 other areas. Studies applying AI to shoulder surgery primarily involved (1) automated imaging analysis including identifying rotator cuff tears and shoulder implants (2) risk prediction analyses including perioperative complications, functional outcomes, and patient satisfaction. Highest model performance area under the curve ranged from 0.681 (poor) to 1.00 (perfect). Only 2 studies reported external validation.

Conclusion

Applications of AI in the field of shoulder surgery are expanding rapidly and offer patient-specific risk stratification for shared decision-making and process automation for resource preservation. However, model performance is modest and external validation remains to be demonstrated, suggesting increased scientific rigor is warranted prior to deploying AI-based clinical applications.

Evaluation of a deep learning method for the automated detection of supraspinatus tears on MRI

OBJECTIVE

To evaluate if deep learning is a feasible approach for automated detection of supraspinatus tears on MRI.

MATERIALS AND METHODS

A total of 200 shoulder MRI studies performed between 2015 and 2019 were retrospectively obtained from our institutional database using a balanced random sampling of studies containing a full-thickness tear, partial-thickness tear, or intact supraspinatus tendon. A 3-stage pipeline was developed comprised of a slice selection network based on a pre-trained residual neural network (ResNet); a segmentation network based on an encoder-decoder network (U-Net); and a custom multi-input convolutional neural network (CNN) classifier. Binary reference labels were created following review of radiologist reports and images by a radiology fellow and consensus validation by two musculoskeletal radiologists. Twenty percent of the data was reserved as a holdout test set with the remaining 80% used for training and optimization under a fivefold cross-validation strategy. Classification and segmentation accuracy were evaluated using area under the receiver operating characteristic curve (AUROC) and Dice similarity coefficient, respectively. Baseline characteristics in correctly versus incorrectly classified cases were compared using independent sample t-test and chi-squared.

RESULTS

Test sensitivity and specificity of the classifier at the optimal Youden's index were 85.0% (95% CI: 62.1-96.8%) and 85.0% (95% CI: 62.1-96.8%), respectively. AUROC was 0.943 (95% CI: 0.820-0.991). Dice segmentation accuracy was 0.814 (95% CI: 0.805-0.826). There was no significant difference in AUROC between 1.5 T and 3.0 T studies. Sub-analysis showed superior sensitivity on full-thickness (100%) versus partial-thickness (72.5%) subgroups.

DATA CONCLUSION

Deep learning is a feasible approach to detect supraspinatus tears on MRI.

Measuring the critical shoulder angle on radiographs: an accurate and repeatable deep learning model

PURPOSE

Since the critical shoulder angle (CSA) is considered a risk factor for shoulder pathology and the intra- and inter-rater variabilities in its calculation are not negligible, we developed a deep learning model that calculates it automatically and accurately.

METHODS

We used a dataset of 8467 anteroposterior x-ray images of the shoulder annotated with 3 landmarks of interest. A Convolutional Neural Network model coupled with a spatial to numerical transform (DSNT) layer was used to predict the landmark coordinates from which the CSA was calculated. The performances were evaluated by calculating the Euclidean distance between the ground truth coordinates and the predicted ones normalized with respect to the distance between the first and the second points, and the error between the CSA angle measured by a human observer and the predicted one.

RESULTS

Regarding the normalized point distances, we obtained a median error of 2.9%, 2.5%, and 2% for the three points among the entire set. Considering CSA calculations, the median errors were 1° (standard deviation 1.2°), 0.88° (standard deviation 0.87°), and 0.99° (standard deviation 1°) for angles below 30°, between 30° and 35°, and above 35°, respectively.

DISCUSSION

These results demonstrate that the model has the potential to be used in clinical settings where the replicability is important. The reported standard error of the CSA measurement is greater than 2° that is above the median error of our model, indicating a potential accuracy sufficient to be used in a clinical setting.

Studying osteoarthritis with artificial intelligence applied to magnetic resonance imaging

The 3D nature and soft-tissue contrast of MRI makes it an invaluable tool for osteoarthritis research, by facilitating the elucidation of disease pathogenesis and progression. The recent increasing employment of MRI has certainly been stimulated by major advances that are due to considerable investment in research, particularly related to artificial intelligence (AI). These AI-related advances are revolutionizing the use of MRI in clinical research by augmenting activities ranging from image acquisition to post-processing. Automation is key to reducing the long acquisition times of MRI, conducting large-scale longitudinal studies and quantitatively defining morphometric and other important clinical features of both soft and hard tissues in various anatomical joints. Deep learning methods have been used recently for multiple applications in the musculoskeletal field to improve understanding of osteoarthritis. Compared with labour-intensive human efforts, AI-based methods have advantages and potential in all stages of imaging, as well as post-processing steps, including aiding diagnosis and prognosis. However, AI-based methods also have limitations, including the arguably limited interpretability of AI models. Given that the AI community is highly invested in uncovering uncertainties associated with model predictions and improving their interpretability, we envision future clinical translation and progressive increase in the use of AI algorithms to support clinicians in optimizing patient care.

3D MRI Models of the Musculoskeletal System

Computed tomography (CT) is most commonly used to produce three-dimensional (3D) models for evaluating bone and joint morphology in clinical practice. However, 3D models created from magnetic resonance imaging (MRI) data can be equally effective for comprehensive and accurate assessment of osseous and soft tissue structure morphology and pathology. The quality of 3D MRI models has steadily increased over time, with growing potential to replace 3D CT models in various musculoskeletal (MSK) applications. In practice, a single MRI examination for two-dimensional and 3D assessments can increase the value of MRI and simplify the pre- and postoperative imaging work-up. Multiple studies have shown excellent performance of 3D MRI models in shoulder injuries, in the hip in the setting of femoroacetabular impingement, and in the knee for the creation of bone surface models. Therefore, the utility of 3D MRI postprocessed models is expected to continue to rise and broaden in applications. Computer-based and artificial intelligence-assisted postprocessing techniques have tremendous potential to improve the efficiency of 3D model creation, opening many research avenues to validate the applicability of 3D MRI and establish 3D-specific quantitative assessment criteria. We provide a practice-focused overview of 3D MRI acquisition strategies, postprocessing techniques for 3D model creation, MSK applications of 3D MRI models, and an illustration of cases from our daily clinical practice.

3D MRI in Osteoarthritis

Osteoarthritis (OA) is among the top 10 burdensome diseases, with the knee the most affected joint. Magnetic resonance imaging (MRI) allows whole-knee assessment, making it ideally suited for imaging OA, considered a multitissue disease. Three-dimensional (3D) MRI enables the comprehensive assessment of OA, including quantitative morphometry of various joint tissues. Manual tissue segmentation on 3D MRI is challenging but may be overcome by advanced automated image analysis methods including artificial intelligence (AI). This review presents examples of the utility of 3D MRI for knee OA, focusing on the articular cartilage, bone, meniscus, synovium, and infrapatellar fat pad, and it highlights several applications of AI that facilitate segmentation, lesion detection, and disease classification.