Deep learning-based screening tool for rotator cuff tears on shoulder radiography

Classification models for arthropathy grades of multiple joints based on hierarchical continual learning

PURPOSE

To develop a hierarchical continual arthropathy classification model for multiple joints that can be updated continuously for large-scale studies of various anatomical structures.

MATERIALS AND METHODS

This study included a total of 1371 radiographs of knee, elbow, ankle, shoulder, and hip joints from three tertiary hospitals. For model development, 934 radiographs of the knee, elbow, ankle, and shoulder were gathered from Sinchon Severance Hospital between July 1 and December 31, 2022. For external validation, 125 hip radiographs were collected from Yongin Severance Hospital between January 1 and December 31, 2022, and 312 knee cases were gathered from Gangnam Severance Hospital between January 1 and June 31, 2023. The Hierarchical Dynamically Expandable Representation (Hi-DER) model was trained stepwise on four joints using five-fold cross-validation. Arthropathy classification was evaluated at three hierarchical levels: abnormal classification (L1), low-grade or high-grade classification (L2), and specific grade classification (L3). The model's performance was compared with the grading predictions of two other AI models and three radiologists. For model explainability, gradient-weighted class activation mapping (Grad-CAM) and progressive erasing plus progressive restoration (PEPPR) were employed.

RESULTS

The model achieved a weighted average AUC of 0.994 (95% CI: 0.985, 0.999) for L1, 0.980 (95% CI: 0.958, 0.996) for L2, and 0.973 (95% CI: 0.943, 0.993) for L3. The model maintained an AUC above 0.800 with 70% of the input regions erased. During external validation on hip joints, the model demonstrated a weighted average AUC of 0.978 (95% CI: 0.952, 0.996) for L1, 0.977 (95% CI: 0.946, 0.996) for L2, and 0.971 (95% CI: 0.934, 0.996) for L3. For external knee data, the model yielded a weighted average AUC of 0.934 (95%: CI 0.904, 0.958), 0.929 (95% CI: 0.900, 0.954), and 0.857 (95% CI: 0.816, 0.894) for L1, L2, and L3, respectively.

CONCLUSION

The Hi-DER may enhance the efficiency of arthropathy diagnosis through accurate classification of arthropathy grades across multiple joints, potentially enabling early treatment.

An accelerated deep learning model can accurately identify clinically important humeral and scapular landmarks on plain radiographs obtained before and after anatomic arthroplasty

PURPOSE

Accurate identification of radiographic landmarks is fundamental to characterizing glenohumeral relationships before and sequentially after shoulder arthroplasty, but manual annotation of these radiographs is laborious. We report on the use of artificial intelligence, specifically computer vision and deep learning models (DLMs), in determining the accuracy of DLM-identified and surgeon identified (SI) landmarks before and after anatomic shoulder arthroplasty.

MATERIALS & METHODS

240 true anteroposterior radiographs were annotated using 11 standard osseous landmarks to train a deep learning model. Radiographs were modified to allow for a training model consisting of 2,260 images. The accuracy of DLM landmarks was compared to manually annotated radiographs using 60 radiographs not used in the training model. In addition, we also performed 14 different measurements of component positioning and compared these to measurements made based on DLM landmarks.

RESULTS

The mean deviation between DLM vs. SI cortical landmarks was 1.9 ± 1.9 mm. Scapular landmarks had slightly lower deviations compared to humeral landmarks (1.5 ± 1.8 mm vs. 2.1 ± 2.0 mm, p < 0.001). The DLM was also found to be accurate with respect to 14 measures of scapular, humeral, and glenohumeral measurements with a mean deviation of 2.9 ± 2.7 mm.

CONCLUSIONS

An accelerated deep learning model using a base of only 240 annotated images was able to achieve low levels of deviation in identifying common humeral and scapular landmarks on preoperative and postoperative radiographs. The reliability and efficiency of this deep learning model represents a powerful tool to analyze preoperative and postoperative radiographs while avoiding human observer bias.

LEVEL OF EVIDENCE

IV.

Medical Imaging Applications Developed Using Artificial Intelligence Demonstrate High Internal Validity Yet Are Limited in Scope and Lack External Validation

PURPOSE

To (1) review definitions and concepts necessary to interpret applications of deep learning (DL; a domain of artificial intelligence that leverages neural networks to make predictions on media inputs such as images) and (2) identify knowledge and translational gaps in the literature to provide insight into specific areas for improvement as adoption of this technology continues.

METHODS

A comprehensive search of the literature was performed in December 2023 for articles regarding the use of DL in sports medicine. For each study, information regarding the joint of focus, specific anatomic structure/pathology to which DL was applied, imaging modality utilized, source of images used for model training and testing, data set size, model performance, and whether the DL model was externally validated was recorded. A numerical scale was used to rate each DL model's clinical impact, with 1 corresponding to proof-of-concept studies with little to no direct clinical impact and 5 corresponding to practice-changing clinical impact and ready for clinical deployment.

RESULTS

Fifty-five studies were identified, all of which were published within the past 5 years, while 82% were published within the past 3 years. Of the DL models identified, 84% were developed for classification tasks, 9% for automated measurements, and 7% for segmentation. A total of 62% of studies utilized magnetic resonance imaging as the imaging modality, 25% radiographs, and 7% ultrasound, while 1 study each used computed tomography, arthroscopic images, or arthroscopic video. Sixty-five percent of studies focused on the detection of tears (anterior cruciate ligament [ACL], rotator cuff [RC], and meniscus). The diagnostic performance of ACL tears, as determined by the area under the receiver operator curve (AUROC), ranged from 0.81 to 0.99 for ACL tears (excellent to near perfect), 0.83 to 0.94 for RC tears (excellent), and from 0.75 to 0.96 for meniscus tears (acceptable to excellent). In addition, 3 studies focused on detection of cartilage lesions had AUROC ranging from 0.90 to 0.92 (excellent performance). However, only 4 (7%) studies externally validated their models, suggesting that they may not be generalizable or may not perform well when applied to populations other than that used to develop the model. Finally, the mean clinical impact score was 2 (range, 1-3) on scale of 1 to 5, corresponding to limited clinical applicability.

CONCLUSIONS

DL models in orthopaedic sports medicine show generally excellent performance (high internal validity) but require external validation to facilitate clinical deployment. In addition, current models have low clinical applicability and fail to advance the field due to a focus on routine tasks and a narrow conceptual framework.

LEVEL OF EVIDENCE

Level IV, scoping review of Level I to IV studies.

European Society for Surgery of the Shoulder and Elbow (SECEC) rotator cuff tear registry Delphi consensus

Background

The purpose of this study was to establish consensus statements via a Delphi process on the factors that should be included in a registry for those patients undergoing rotator cuff tear treatment.

Methods

A consensus process on the treatment of rotator cuff utilizing a modified Delphi technique was conducted. Fifty-seven surgeons completed these consensus statements and 9 surgeons declined. The participants were members of the European Society for Surgery of the Shoulder and Elbow committees representing 23 European countries. Thirteen questions were generated regarding the diagnosis and follow-up of rotator cuff tears were distributed, with 3 rounds of questionnaires and final voting occurring. Consensus was defined as achieving 80%-89% agreement, whereas strong consensus was defined as 90%-99% agreement, and unanimous consensus was defined by 100% agreement with a proposed statement.

Results

Of the 13 total questions and consensus statements on rotator cuff tears, 1 achieved unanimous consensus, 6 achieved strong consensus, 5 achieved consensus, and 1 did not achieve consensus. The statement that reached unanimous consensus was that the factors in the patient history that should be evaluated and recorded in the setting of suspected/known rotator cuff tear are age, gender, comorbidities, smoking, traumatic etiology, prior treatment including physical therapy/injections, pain, sleep disturbance, sports, occupation, workmen's compensation, hand dominance, and functional limitations. The statement that did not achieve consensus was related to the role of ultrasound in the initial diagnosis of patients with rotator cuff tears.

Conclusion

Nearly all questions reached consensus among 57 European Society for Surgery of the Shoulder and Elbow members representing 23 different European countries. We encourage surgeons to use this minimum set of variables to establish rotator cuff registries and multicenter studies. By adapting and using compatible variables, data can more easily be compared and eventually merged across countries.

Advancements in the diagnosis and management of rotator cuff tears. The role of artificial intelligence

Background

This review examined the role of artificial intelligence (AI) in the diagnosis and management of rotator cuff tears (RCTs).

Methods

A literature search was conducted in October 2023 using PubMed (MEDLINE), SCOPUS, and EMBASE databases, included only peer-reviewed studies. Relevant articles on AI technology in RCTs. A critical analysis of the relevant literature was conducted.

Results

AI is transforming RCTs management through faster and more precise identification and assessment using algorithms that facilitate segmentation, quantification, and classification of the RCTs across various imaging modalities. Precise algorithms focusing on preoperative factors to assess RCTs reparability have been developed for personalized treatment planning and outcome prediction. AI also aids in exercise classification and promotes patient adherence during at-home physiotherapy. Despite promising advancements, challenges in data quality and symptom integration persist. Future research should include refining AI algorithms, expanding their integration into various imaging techniques, and exploring their roles in postoperative care and surgical decision-making.

Conclusions

AI-driven solutions improve diagnostic accuracy and have the potential to influence treatment planning and postoperative outcomes through the automated RCTs analysis of medical imaging. Integration of high-quality datasets and clinical symptoms into AI models can enhance their reliability. Current AI algorithms can also be refined, integrated into other imaging techniques, and explored further in surgical decision-making and postoperative care.

Artificial Intelligence and Machine Learning in Rotator Cuff Tears

Rotator cuff tears (RCTs) negatively impacts patient well-being. Artificial intelligence (AI) is emerging as a promising tool in medical decision-making. Within AI, deep learning allows to autonomously solve complex tasks. This review assesses the current and potential applications of AI in the management of RCT, focusing on diagnostic utility, challenges, and future perspectives. AI demonstrates promise in RCT diagnosis, aiding clinicians in interpreting complex imaging data. Deep learning frameworks, particularly convoluted neural networks architectures, exhibit remarkable diagnostic accuracy in detecting RCTs on magnetic resonance imaging. Advanced segmentation algorithms improve anatomic visualization and surgical planning. AI-assisted radiograph interpretation proves effective in ruling out full-thickness tears. Machine learning models predict RCT diagnosis and postoperative outcomes, enhancing personalized patient care. Challenges include small data sets and classification complexities, especially for partial thickness tears. Current applications of AI in RCT management are promising yet experimental. The potential of AI to revolutionize personalized, efficient, and accurate care for RCT patients is evident. The integration of AI with clinical expertise holds potential to redefine treatment strategies and optimize patient outcomes. Further research, larger data sets, and collaborative efforts are essential to unlock the transformative impact of AI in orthopedic surgery and RCT management.