Diagnostic Performance of Artificial Intelligence for Detection of Scaphoid and Distal Radius Fractures: A Systematic Review

PURPOSE

To review the existing literature to (1) determine the diagnostic efficacy of artificial intelligence (AI) models for detecting scaphoid and distal radius fractures and (2) compare the efficacy to human clinical experts.

METHODS

PubMed, OVID/Medline, and Cochrane libraries were queried for studies investigating the development, validation, and analysis of AI for the detection of scaphoid or distal radius fractures. Data regarding study design, AI model development and architecture, prediction accuracy/area under the receiver operator characteristic curve (AUROC), and imaging modalities were recorded.

RESULTS

A total of 21 studies were identified, of which 12 (57.1%) used AI to detect fractures of the distal radius, and nine (42.9%) used AI to detect fractures of the scaphoid. AI models demonstrated good diagnostic performance on average, with AUROC values ranging from 0.77 to 0.96 for scaphoid fractures and from 0.90 to 0.99 for distal radius fractures. Accuracy of AI models ranged between 72.0% to 90.3% and 89.0% to 98.0% for scaphoid and distal radius fractures, respectively. When compared to clinical experts, 13 of 14 (92.9%) studies reported that AI models demonstrated comparable or better performance. The type of fracture influenced model performance, with worse overall performance on occult scaphoid fractures; however, models trained specifically on occult fractures demonstrated substantially improved performance when compared to humans.

CONCLUSIONS

AI models demonstrated excellent performance for detecting scaphoid and distal radius fractures, with the majority demonstrating comparable or better performance compared with human experts. Worse performance was demonstrated on occult fractures. However, when trained specifically on difficult fracture patterns, AI models demonstrated improved performance.

CLINICAL RELEVANCE

AI models can help detect commonly missed occult fractures while enhancing workflow efficiency for distal radius and scaphoid fracture diagnoses. As performance varies based on fracture type, future studies focused on wrist fracture detection should clearly define whether the goal is to (1) identify difficult-to-detect fractures or (2) improve workflow efficiency by assisting in routine tasks.

Biomechanical Characteristics of Ulnar Collateral Ligament Injuries Treated With and Without Augmentation: A Network Meta-analysis of Controlled Laboratory Studies

BACKGROUND

Treatment of ulnar collateral ligament (UCL) tears with suture tape augmentation has gained interest given preliminary reports of favorable biomechanical characteristics. No study to date has quantitatively assessed the biomechanical effects of multiple augmentation techniques relative to the native UCL.

PURPOSE

To perform a systematic review and meta-analysis of controlled laboratory studies to assess and comparatively rank biomechanical effects of UCL repair or reconstruction with or without augmentation.

STUDY DESIGN

Systematic review and meta-analysis; Level of evidence, 4.

METHODS

PubMed, OVID/Medline, and Cochrane databases were queried in January 2023. A frequentist network meta-analytic approach was used to perform mixed-treatment comparisons of UCL repair and reconstruction techniques with and without augmentation, with the native UCL as the reference condition. Pooled treatment estimates were quantified under the random-effects assumption. Competing treatments were ranked in the network meta-analysis by using point estimates and standard errors to calculate P scores (greater P score indicates superiority of treatment for given outcome).

RESULTS

Ten studies involving 206 elbow specimens in which a distal UCL tear was simulated were included. UCL reconstruction with suture tape augmentation (AugRecon) restored load to failure to a statistically noninferior magnitude (mean difference [MD], -1.99 N·m; 95% CI, -10.2 to 6.2 N·m; P = .63) compared with the native UCL. UCL reconstruction (Recon) (MD, -12.7 N·m; P < .001) and UCL repair with suture tape augmentation (AugRepair) (MD, -14.8 N·m; P < .001) were both statistically inferior to the native UCL. The AugRecon condition conferred greater load to failure compared with Recon (P < .001) and AugRepair (P = .002) conditions. AugRecon conferred greater torsional stiffness relative to all other conditions and was not statistically different from the native UCL (MD, 0.32 N·m/deg; 95% CI, -0.30 to 0.95 N·m/deg; P = .31). Medial ulnohumeral gapping was not statistically different for the AugRepair (MD, 0.30 mm; 95% CI, -1.22 to 1.82 mm; P = .70), AugRecon (MD, 0.57 mm; 95% CI, -0.70 to 1.84 mm; P = .38), or Recon (MD, 1.02 mm; 95% CI, -0.02 to 2.05 mm; P = .055) conditions compared with the native UCL. P-score analysis indicated that AugRecon was the most effective treatment for increasing ultimate load to failure and torsional stiffness, whereas AugRepair was the most effective for minimizing medial gapping.

CONCLUSION

AugRecon restored load to failure and torsional stiffness most similar to the parameters of the native UCL, whereas Recon and AugRepair did not restore the same advantageous properties at time zero. Medial ulnohumeral gapping during a valgus load was minimized by all 3 treatments. Based on network interactions, AugRecon was the superior treatment approach for restoring important biomechanical features of the UCL at time zero that are jeopardized during a complete distal tear.

A Machine Learning Model Demonstrates Excellent Performance in Predicting Subscapularis Tears Based on Pre-Operative Imaging Parameters Alone

PURPOSE

To develop a machine learning model capable of identifying subscapularis tears before surgery based on imaging and physical examination findings.

METHODS

Between 2010 and 2020, 202 consecutive shoulders underwent arthroscopic rotator cuff repair by a single surgeon. Patient demographics, physical examination findings (including range of motion, weakness with internal rotation, lift/push-off test, belly press test, and bear hug test), and imaging (including direct and indirect signs of tearing, biceps status, fatty atrophy, cystic changes, and other similar findings) were included for model creation.

RESULTS

Sixty percent of the shoulders had partial or full thickness tears of the subscapularis verified during surgery (83% of these were upper third). Using only preoperative imaging-related parameters, the XGBoost model demonstrated excellent performance at predicting subscapularis tears (c-statistic, 0.84; accuracy, 0.85; F1 score, 0.87). The top 5 features included direct signs related to the presence of tearing as evidenced on magnetic resonance imaging (MRI) (changes in tendon morphology and signal), as well as the quality of the MRI and biceps pathology.

CONCLUSIONS

In this study, machine learning was successful in predicting subscapularis tears by MRI alone in 85% of patients, and this accuracy did not decrease by isolating the model to the top features. The top five features included direct signs related to the presence of tearing as evidenced on MRI (changes in tendon morphology and signal), as well as the quality of the MRI and biceps pathology. Last, in advanced modeling, the addition of physical examination or patient characteristics did not make a significant difference in the predictive ability of this model.

LEVEL OF EVIDENCE

Level III, diagnostic case-control study.

A practical guide to the development and deployment of deep learning models for the orthopaedic surgeon: Part III, focus on registry creation, diagnosis, and data privacy

Deep learning is a subset of artificial intelligence (AI) with enormous potential to transform orthopaedic surgery. As has already become evident with the deployment of Large Language Models (LLMs) like ChatGPT (OpenAI Inc.), deep learning can rapidly enter clinical and surgical practices. As such, it is imperative that orthopaedic surgeons acquire a deeper understanding of the technical terminology, capabilities and limitations associated with deep learning models. The focus of this series thus far has been providing surgeons with an overview of the steps needed to implement a deep learning-based pipeline, emphasizing some of the important technical details for surgeons to understand as they encounter, evaluate or lead deep learning projects. However, this series would be remiss without providing practical examples of how deep learning models have begun to be deployed and highlighting the areas where the authors feel deep learning may have the most profound potential. While computer vision applications of deep learning were the focus of Parts I and II, due to the enormous impact that natural language processing (NLP) has had in recent months, NLP-based deep learning models are also discussed in this final part of the series. In this review, three applications that the authors believe can be impacted the most by deep learning but with which many surgeons may not be familiar are discussed: (1) registry construction, (2) diagnostic AI and (3) data privacy. Deep learning-based registry construction will be essential for the development of more impactful clinical applications, with diagnostic AI being one of those applications likely to augment clinical decision-making in the near future. As the applications of deep learning continue to grow, the protection of patient information will become increasingly essential; as such, applications of deep learning to enhance data privacy are likely to become more important than ever before. Level of Evidence: Level IV.

Unsupervised Machine Learning of the Combined Danish and Norwegian Knee Ligament Registers: Identification of 5 Distinct Patient Groups With Differing ACL Revision Rates

BACKGROUND

Most clinical machine learning applications use a supervised learning approach using labeled variables. In contrast, unsupervised learning enables pattern detection without a prespecified outcome.

PURPOSE/HYPOTHESIS

The purpose of this study was to apply unsupervised learning to the combined Danish and Norwegian knee ligament register (KLR) with the goal of detecting distinct subgroups. It was hypothesized that resulting groups would have differing rates of subsequent anterior cruciate ligament reconstruction (ACLR) revision.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

K-prototypes clustering was performed on the complete case KLR data. After performing the unsupervised learning analysis, the authors defined clinically relevant characteristics of each cluster using variable summaries, surgeons' domain knowledge, and Shapley Additive exPlanations analysis.

RESULTS

Five clusters were identified. Cluster 1 (revision rate, 9.9%) patients were young (mean age, 22 years; SD, 6 years), received hamstring tendon (HT) autograft (91%), and had lower baseline Knee injury and Osteoarthritis Outcome Score (KOOS) Sport and Recreation (Sports) scores (mean, 25.0; SD, 15.6). Cluster 2 (revision rate, 6.9%) patients received HT autograft (89%) and had higher baseline KOOS Sports scores (mean, 67.2; SD, 16.5). Cluster 3 (revision rate, 4.7%) patients received bone-patellar tendon-bone (BPTB) or quadriceps tendon (QT) autograft (94%) and had higher baseline KOOS Sports scores (mean, 65.8; SD, 16.4). Cluster 4 (revision rate, 4.1%) patients received BPTB or QT autograft (88%) and had low baseline KOOS Sports scores (mean, 20.5; SD, 14.0). Cluster 5 (revision rate, 3.1%) patients were older (mean age, 42 years; SD, 7 years), received HT autograft (89%), and had low baseline KOOS Sports scores (mean, 23.4; SD, 17.6).

CONCLUSION

Unsupervised learning identified 5 distinct KLR patient subgroups and each grouping was associated with a unique ACLR revision rate. Patients can be approximately classified into 1 of the 5 clusters based on only 3 variables: age, graft choice (HT, BPTB, or QT autograft), and preoperative KOOS Sports subscale score. If externally validated, the resulting groupings may enable quick risk stratification for future patients undergoing ACLR in the clinical setting. Patients in cluster 1 are considered high risk (9.9%), cluster 2 patients medium risk (6.9%), and patients in clusters 3 to 5 low risk (3.1%-4.7%) for revision ACLR.

Platelet-Rich Plasma Versus Alternative Injections for Osteoarthritis of the Knee: A Systematic Review and Statistical Fragility Index-Based Meta-analysis of Randomized Controlled Trials

BACKGROUND

Based in part on the results of randomized controlled trials (RCTs) that suggest a beneficial effect over alternative treatment options, the use of platelet-rich plasma (PRP) for the management of knee osteoarthritis (OA) is widespread and increasing. However, the extent to which these studies are vulnerable to slight variations in the outcomes of patients remains unknown.

PURPOSE

To evaluate the statistical fragility of conclusions from RCTs that reported outcomes of patients with knee OA who were treated with PRP versus alternative nonoperative management strategies.

STUDY DESIGN

Systematic review and meta-analysis; Level of evidence, 2.

METHODS

All RCTs comparing PRP with alternative nonoperative treatment options for knee OA were identified. The fragility index (FI) and reverse FI were applied to assess the robustness of conclusions regarding the efficacy of PRP for knee OA. Meta-analyses were performed to determine the minimum number of patients from ≥1 trials included in the meta-analysis for which a modification on the event status would change the statistical significance of the pooled treatment effect.

RESULTS

In total, this analysis included outcomes from 1993 patients with a mean ± SD age of 58.0 ± 3.8 years. The mean number of events required to reverse significance of individual RCTs (FI) was 4.57 ± 5.85. Based on random-effects meta-analyses, PRP demonstrated a significantly higher rate of successful outcomes when compared with hyaluronic acid (P = .002; odds ratio [OR], 2.19; 95% CI, 1.33-3.62), as well as higher rates of patient-reported symptom relief (P = .019; OR, 1.55; 95% CI, 1.07-2.24), not requiring a reintervention after the initial injection treatment (P = .002; OR, 2.17; 95% CI, 1.33-3.53), and achieving the minimal clinically important difference (MCID) for pain improvement (P = .007; OR, 6.19; 95% CI, 1.63-23.42) when compared with all alternative nonoperative treatments. Overall, the mean number of events per meta-analysis required to change the statistical significance of the pooled treatment effect was 8.67 ± 4.50.

CONCLUSION

Conclusions drawn from individual RCTs evaluating PRP for knee OA demonstrated slight robustness. On meta-analysis, PRP demonstrated a significant advantage over hyaluronic acid as well as improved symptom relief, lower rates of reintervention, and more frequent achievement of the MCID for pain improvement when compared with alternative nonoperative treatment options. Statistically significant pooled treatment effects evaluating PRP for knee OA are more robust than approximately half of all comparable meta-analyses in medicine and health care. Future RCTs and meta-analyses should consider reporting FIs and fragility quotients to facilitate interpretation of results in their proper context.

Machine Learning/Artificial Intelligence in Sports Medicine: State of the Art and Future Directions

Machine learning (ML) is changing the way health care is practiced and recent applications of these novel statistical techniques have started to impact orthopaedic sports medicine. Machine learning enables the analysis of large volumes of data to establish complex relationships between "input" and "output" variables. These relationships may be more complex than could be established through traditional statistical analysis and can lead to the ability to predict the "output" with high levels of accuracy. Supervised learning is the most common ML approach for healthcare data and recent studies have developed algorithms to predict patient-specific outcome after surgical procedures such as hip arthroscopy and anterior cruciate ligament reconstruction. Deep learning is a higher-level ML approach that facilitates the processing and interpretation of complex datasets through artificial neural networks that are inspired by the way the human brain processes information. In orthopaedic sports medicine, deep learning has primarily been used for automatic image (computer vision) and text (natural language processing) interpretation. While applications in orthopaedic sports medicine have been increasing exponentially, one significant barrier to widespread adoption of ML remains clinician unfamiliarity with the associated methods and concepts. The goal of this review is to introduce these concepts, review current machine learning models in orthopaedic sport medicine, and discuss future opportunities for innovation within the specialty.

External validation of the Norwegian anterior cruciate ligament reconstruction revision prediction model using patients from the STABILITY 1 Trial

PURPOSE

A machine learning-based anterior cruciate ligament (ACL) revision prediction model has been developed using Norwegian Knee Ligament Register (NKLR) data, but lacks external validation outside Scandinavia. This study aimed to assess the external validity of the NKLR model (https://swastvedt.shinyapps.io/calculator_rev/) using the STABILITY 1 randomized clinical trial (RCT) data set. The hypothesis was that model performance would be similar.

METHODS

The NKLR Cox Lasso model was selected for external validation owing to its superior performance in the original study. STABILITY 1 patients with all five predictors required by the Cox Lasso model were included. The STABILITY 1 RCT was a prospective study which randomized patients to receive either a hamstring tendon autograft (HT) alone or HT plus a lateral extra-articular tenodesis (LET). Since all patients in the STABILITY 1 trial received HT ± LET, three configurations were tested: 1: all patients coded as HT, 2: HT + LET group coded as bone-patellar tendon-bone (BPTB) autograft, 3: HT + LET group coded as unknown/other graft choice. Model performance was assessed via concordance and calibration.

RESULTS

In total, 591/618 (95.6%) STABILITY 1 patients were eligible for inclusion, with 39 undergoing revisions within 2 years (6.6%). Model performance was best when patients receiving HT + LET were coded as BPTB. Concordance was similar to the original NKLR prediction model for 1- and 2-year revision prediction (STABILITY: 0.71; NKLR: 0.68-0.69). Concordance 95% confidence interval (CI) ranged from 0.63 to 0.79. The model was well calibrated for 1-year prediction while the 2-year prediction demonstrated evidence of miscalibration.

CONCLUSION

When patients in STABILITY 1 who received HT + LET were coded as BPTB in the NKLR prediction model, concordance was similar to the index study. However, due to a wide 95% CI, the true performance of the prediction model with this Canadian and European cohort is unclear and a larger data set is required to definitively determine the external validity. Further, better calibration for 1-year predictions aligns with general prediction modelling challenges over longer periods. While not a large enough sample size to elicit the true accuracy and external validity of the prediction model when applied to North American patients, this analysis provides more support for the notion that HT plus LET performs similarly to BPTB reconstruction. In addition, despite the wide confidence interval, this study suggests optimism regarding the accuracy of the model when applied outside of Scandinavia.

LEVEL OF EVIDENCE

Level 3, cohort study.

Risk of Arthroplasty in Patients with Subchondral Insufficiency Fractures of the Knee: A Matched Study of the Implantable Shock Absorber using a Validated Predictive Model

Subchondral insufficiency fractures of the knee (SIFK) can result in high rates of osteoarthritis and arthroplasty. The implantable shock absorber (ISA) implant is a titanium and polycarbonate urethane device which reduces the load on the medial compartment of the knee by acting as an extra-articular load absorber while preserving the joint itself. The purpose of this study was to evaluate whether partially unloading the knee with the ISA altered the likelihood of progression to arthroplasty utilizing a validated predictive risk model (SIFK score). A retrospective case-control (2:1) study was performed on patients with SIFK without any previous surgery and on those implanted with the ISA with the primary outcome being progression to arthroplasty compared with nonoperative treatment at 2 years. Baseline and final radiographs, as well as magnetic resonance imagings, were reviewed for the evaluation of meniscus or ligament injuries, insufficiency fractures, and subchondral edema. Patients from a prospective study were matched using the exact SIFK Score, a validated predictive score for progression to arthroplasty in patients with SIFK, to those who received the ISA implant. Kaplan-Meier analysis was conducted to assess survival. A total of 57 patients (38 controls:19 ISA) with a mean age of 60.6 years and 54% female were included. The SIFK score was matched exactly between cases and controls for all patients. The 2-year survival rate of 100% for the ISA group was significantly higher than the corresponding rate of 61% for the control group (p < 0.01). In ISA, 0% of the patients converted to arthroplasty at 2 years, and 5% (one patient) had hardware removal at 1 year. When stratified by risk, the ISA group did not have a significantly higher survival compared with low-risk (p = 0.3) or medium-risk (p = 0.2) controls, though it had a significantly higher survival for high-risk groups at 2 years (100 vs. 15%, p < 0.01). SIFK of the medial knee can lead to significant functional limitation and high rates of conversion to arthroplasty. Implants such as the ISA have the potential to alter the progression to arthroplasty in these patients, especially those at high risk.

Autologous Osteochondral Transfer of the Knee Demonstrates Continued High Rates of Return to Sport and Low Rates of Conversion to Arthroplasty at Long-Term Follow-Up: A Systematic Review

PURPOSE

To perform a systematic review of the literature to evaluate (1) activity level and knee function, (2) reoperation and failure rates, and (3) risk factors for reoperation and failure of autologous osteochondral transfer (AOT) at long-term follow-up.

METHODS

A comprehensive review of the long-term outcomes of AOT was performed. Studies reported on activity-based outcomes (Tegner Activity Scale) and clinical outcomes (Lysholm score and International Knee Documentation Committee score). Reoperation and failure rates as defined by the publishing authors were recorded for each study. Modified Coleman Methodology Scores were calculated to assess study methodological quality.

RESULTS

Twelve studies with a total of 495 patients and an average age of 32.5 years at the time of surgery and a mean follow-up of 15.1 years (range, 10.4-18.0 years) were included. The mean defect size was 3.2 cm2 (range, 1.9-6.9 cm2). The mean duration of symptoms before surgery was 5.1 years. Return to sport rates ranged from 86% to 100%. Conversion to arthroplasty rates ranged from 0% to 16%. The average preoperative International Knee Documentation Committee scores ranged from 32.9 to 36.8, and the average postoperative International Knee Documentation Committee scores at final follow-up ranged from 66.3 to 77.3. The average preoperative Lysholm scores ranged from 44.5 to 56.0 and the average postoperative Lysholm scores ranged from 70.0 to 96.5. The average preoperative Tegner scores ranged from 2.5 to 3.0, and the average postoperative scores ranged from 4.1 to 7.0.

CONCLUSIONS

AOT of the knee resulted in high rates of return to sport with correspondingly low rates of conversion to arthroplasty at long-term follow-up. In addition, AOT demonstrated significant improvements in long-term patient-reported outcomes from baseline.

LEVEL OF EVIDENCE

Level IV, systematic review of Level I-IV studies.

Deep Learning Artificial Intelligence Tool for Automated Radiographic Determination of Posterior Tibial Slope in Patients With ACL Injury

Background

An increased posterior tibial slope (PTS) corresponds with an increased risk of graft failure after anterior cruciate ligament (ACL) reconstruction (ACLR). Validated methods of manual PTS measurements are subject to potential interobserver variability and can be inefficient on large datasets.

Purpose/Hypothesis

To develop a deep learning artificial intelligence technique for automated PTS measurement from standard lateral knee radiographs. It was hypothesized that this deep learning tool would be able to measure the PTS on a high volume of radiographs expeditiously and that these measurements would be similar to previously validated manual measurements.

Study Design

Cohort study (diagnosis); Level of evidence, 2.

Methods

A deep learning U-Net model was developed on a cohort of 300 postoperative short-leg lateral radiographs from patients who underwent ACLR to segment the tibial shaft, tibial joint surface, and tibial tuberosity. The model was trained via a random split after an 80 to 20 train-validation scheme. Masks for training images were manually segmented, and the model was trained for 400 epochs. An image processing pipeline was then deployed to annotate and measure the PTS using the predicted segmentation masks. Finally, the performance of this combined pipeline was compared with human measurements performed by 2 study personnel using a previously validated manual technique for measuring the PTS on short-leg lateral radiographs on an independent test set consisting of both pre- and postoperative images.

Results

The U-Net semantic segmentation model achieved a mean Dice similarity coefficient of 0.885 on the validation cohort. The mean difference between the human-made and computer-vision measurements was 1.92° (σ = 2.81° [P = .24]). Extreme disagreements between the human and machine measurements, as defined by ≥5° differences, occurred <5% of the time. The model was incorporated into a web-based digital application front-end for demonstration purposes, which can measure a single uploaded image in Portable Network Graphics format in a mean time of 5 seconds.

Conclusion

We developed an efficient and reliable deep learning computer vision algorithm to automate the PTS measurement on short-leg lateral knee radiographs. This tool, which demonstrated good agreement with human annotations, represents an effective clinical adjunct for measuring the PTS as part of the preoperative assessment of patients with ACL injuries.

Current clinical applications of artificial intelligence in shoulder surgery: what the busy shoulder surgeon needs to know and what's coming next

Background

Artificial intelligence (AI) is a continuously expanding field with the potential to transform a variety of industries-including health care-by providing automation, efficiency, precision, accuracy, and decision-making support for simple and complex tasks. Basic knowledge of the key features as well as limitations of AI is paramount to understand current developments in this field and to successfully apply them to shoulder surgery. The purpose of the present review is to provide an overview of AI within orthopedics and shoulder surgery exploring current and forthcoming AI applications.

Methods

PubMed and Scopus databases were searched to provide a narrative review of the most relevant literature on AI applications in shoulder surgery.

Results

Despite the enormous clinical and research potential of AI, orthopedic surgery has been a relatively late adopter of AI technologies. Image evaluation, surgical planning, aiding decision-making, and facilitating patient evaluations over time are some of the current areas of development with enormous opportunities to improve surgical practice, research, and education. Furthermore, the advancement of AI-driven strategies has the potential to create a more efficient medical system that may reduce the overall cost of delivering and implementing quality health care for patients with shoulder pathology.

Conclusion

AI is an expanding field with the potential for broad clinical and research applications in orthopedic surgery. Many challenges still need to be addressed to fully leverage the potential of AI to clinical practice and research such as privacy issues, data ownership, and external validation of the proposed models.

Artificial intelligence for automated identification of total shoulder arthroplasty implants

BACKGROUND

Accurate and rapid identification of implant manufacturer and model is critical in the evaluation and management of patients requiring revision total shoulder arthroplasty (TSA). Failure to correctly identify implant designs in these circumstances may lead to delay in care, unexpected intraoperative challenges, increased morbidity, and excess health care costs. Deep learning (DL) permits automated image processing and holds the potential to mitigate such challenges while improving the value of care rendered. The purpose of this study was to develop an automated DL algorithm to identify shoulder arthroplasty implants from plain radiographs.

METHODS

A total of 3060 postoperative images from patients who underwent TSA between 2011 and 2021 performed by 26 fellowship-trained surgeons at 2 independent tertiary academic hospitals in the Pacific Northwest and Mid-Atlantic Northeast were included. A DL algorithm was trained using transfer learning and data augmentation to classify 22 different reverse TSA and anatomic TSA prostheses from 8 implant manufacturers. Images were split into training and testing cohorts (2448 training and 612 testing images). Optimized model performance was assessed using standardized metrics including the multiclass area under the receiver operating characteristic curve (AUROC) and compared with a reference standard of implant data from operative reports.

RESULTS

The algorithm classified implants at a mean speed of 0.079 seconds (±0.002 seconds) per image. The optimized model discriminated between 8 manufacturers (22 unique implants) with AUROCs of 0.994-1.000, accuracy of 97.1%, and sensitivities between 0.80 and 1.00 on the independent testing set. In the subset of single-institution implant predictions, a DL model identified 6 specific implants with AUROCs of 0.999-1.000, accuracy of 99.4%, and sensitivity >0.97 for all implants. Saliency maps revealed key differentiating features across implant manufacturers and designs recognized by the algorithm for classification.

CONCLUSION

A DL model demonstrated excellent accuracy in identifying 22 unique TSA implants from 8 manufacturers. This algorithm may provide a clinically meaningful adjunct in assisting with preoperative planning for the failed TSA and allows for scalable expansion with additional radiographic data and validation efforts.

Application of Machine Learning Algorithms for Prognostic Assessment in Rotator Cuff Pathologies: A Clinical Data-Based Approach

AIM

The overall aim of this proposal is to ameliorate the care of rotator cuff (RC) tear patients by applying an innovative machine learning approach for outcome prediction after arthroscopic repair.

MATERIALS AND METHODS

We applied state-of-the-art machine learning algorithms to evaluate the best predictors of the outcome, and 100 RC patients were evaluated at baseline (T0), after 1 month (T1), 3 months (T2), 6 months (T3), and 1 year (T4) from surgical intervention. The outcome measure was the Costant-Murley Shoulder Score, whereas age, sex, BMI, the 36-Item Short-Form Survey, the Simple Shoulder Test, the Hospital Anxiety and Depression Scale, the American Shoulder and Elbow Surgeons Score, the Oxford Shoulder Score, and the Shoulder Pain and Disability Index were considered as predictive factors. Support vector machine (SVM), k-nearest neighbors (k-NN), naïve Bayes (NB), and random forest (RF) algorithms were employed.

RESULTS

Across all sessions, the classifiers demonstrated suboptimal performance when using both the complete and shrunken sets of features. Specifically, the logistic regression (LR) classifier achieved a mean accuracy of 46.5% ± 6%, while the random forest (RF) classifier achieved 51.25% ± 4%. For the shrunken set of features, LR obtained a mean accuracy of 48.5% ± 6%, and RF achieved 45.5% ± 4.5%. No statistical differences were found when comparing the performance metrics of ML algorithms.

CONCLUSIONS

This study underlines the importance of extending the application of AI methods to new predictors, such as neuroimaging and kinematic data, in order to better record significant shifts in RC patients' prognosis.

LIMITATIONS

The data quality within the cohort could represent a limitation, since certain variables, such as smoking, diabetes, and work injury, are known to have an impact on the outcome.

Understanding risk for early dislocation resulting in reoperation within 90 days of reverse total shoulder arthroplasty: extreme rare event detection through cost-sensitive machine learning

BACKGROUND

Reliable prediction of postoperative dislocation after reverse total shoulder arthroplasty (RSA) would inform patient counseling as well as surgical and postoperative decision making. Understanding interactions between multiple risk factors is important to identify those patients most at risk of this rare but costly complication. To better understand these interactions, a game theory-based approach was undertaken to develop machine learning models capable of predicting dislocation-related 90-day readmission following RSA.

MATERIAL & METHODS

A retrospective review of the Nationwide Readmissions Database was performed to identify patients who underwent RSA between 2016 and 2018 with a subsequent readmission for prosthetic dislocation. Of the 74,697 index procedures included in the data set, 740 (1%) experienced a dislocation resulting in hospital readmission within 90 days. Five machine learning algorithms were evaluated for their ability to predict dislocation leading to hospital readmission within 90 days of RSA. Shapley additive explanation (SHAP) values were calculated for the top-performing models to quantify the importance of features and understand variable interaction effects, with hierarchical clustering used to identify cohorts of patients with similar risk factor combinations.

RESULTS

Of the 5 models evaluated, the extreme gradient boosting algorithm was the most reliable in predicting dislocation (C statistic = 0.71, F2 score = 0.07, recall = 0.84, Brier score = 0.21). SHAP value analysis revealed multifactorial explanations for dislocation risk, with presence of a preoperative humerus fracture; disposition involving discharge or transfer to a skilled nursing facility, intermediate care facility, or other nonroutine facility; and Medicaid as the expected primary payer resulting in strong, positive, and unidirectional effects on increasing dislocation risk. In contrast, factors such as comorbidity burden, index procedure complexity and duration, age, sex, and presence or absence of preoperative glenohumeral osteoarthritis displayed bidirectional influences on risk, indicating potential protective effects for these variables and opportunities for risk mitigation. Hierarchical clustering using SHAP values identified patients with similar risk factor combinations.

CONCLUSION

Machine learning can reliably predict patients at risk for postoperative dislocation resulting in hospital readmission within 90 days of RSA. Although individual risk for dislocation varies significantly based on unique combinations of patient characteristics, SHAP analysis revealed a particularly at-risk cohort consisting of young, male patients with high comorbidity burdens who are indicated for RSA after a humerus fracture. These patients may require additional modifications in postoperative activity, physical therapy, and counseling on risk-reducing measures to prevent early dislocation after RSA.

Ceiling Effect of the Combined Norwegian and Danish Knee Ligament Registers Limits Anterior Cruciate Ligament Reconstruction Outcome Prediction

BACKGROUND

Clinical tools based on machine learning analysis now exist for outcome prediction after primary anterior cruciate ligament reconstruction (ACLR). Relying partly on data volume, the general principle is that more data may lead to improved model accuracy.

PURPOSE/HYPOTHESIS

The purpose was to apply machine learning to a combined data set from the Norwegian and Danish knee ligament registers (NKLR and DKRR, respectively), with the aim of producing an algorithm that can predict revision surgery with improved accuracy relative to a previously published model developed using only the NKLR. The hypothesis was that the additional patient data would result in an algorithm that is more accurate.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

Machine learning analysis was performed on combined data from the NKLR and DKRR. The primary outcome was the probability of revision ACLR within 1, 2, and 5 years. Data were split randomly into training sets (75%) and test sets (25%). There were 4 machine learning models examined: Cox lasso, random survival forest, gradient boosting, and super learner. Concordance and calibration were calculated for all 4 models.

RESULTS

The data set included 62,955 patients in which 5% underwent a revision surgical procedure with a mean follow-up of 7.6 ± 4.5 years. The 3 nonparametric models (random survival forest, gradient boosting, and super learner) performed best, demonstrating moderate concordance (0.67 [95% CI, 0.64-0.70]), and were well calibrated at 1 and 2 years. Model performance was similar to that of the previously published model (NKLR-only model: concordance, 0.67-0.69; well calibrated).

CONCLUSION

Machine learning analysis of the combined NKLR and DKRR enabled prediction of the revision ACLR risk with moderate accuracy. However, the resulting algorithms were less user-friendly and did not demonstrate superior accuracy in comparison with the previously developed model based on patients from the NKLR alone, despite the analysis of nearly 63,000 patients. This ceiling effect suggests that simply adding more patients to current national knee ligament registers is unlikely to improve predictive capability and may prompt future changes to increase variable inclusion.

Improved 2-Year Freedom from Arthroplasty in Patients with High-Risk SIFK Scores and Medial Knee Osteoarthritis Treated with an Implantable Shock Absorber versus Non-Operative Care

OBJECTIVE

Subchondral insufficiency fracture of the knee (SIFK) is associated with high rates of osteoarthritis (OA) and arthroplasty. The implantable shock absorber (ISA) is an extra-capsular implant that unloads the medial knee compartment. This study compared the 2-year freedom from arthroplasty rates in subjects with medial knee OA and SIFK when treated with an ISA versus a matched cohort of patients treated non-surgically.

DESIGN

This retrospective case-control study compared 2-year conversion rates to arthroplasty in SIFK score-, age-, and body mass index (BMI)-matched control subjects without prior surgical history with ISA-implanted subjects from an ongoing prospective study. Baseline and final radiographs, and MRIs were reviewed for evaluation of meniscus or ligament injuries, insufficiency fractures, and subchondral edema. Kaplan-Meier analysis assessed survival.

RESULTS

Forty-two patients (21 Control: 21 ISA), mean age = 52.3 ± 8.7 years, BMI = 29.5 ± 3.9 kg/m2, 40% female were evaluated. Both ISA and Control arms had the same numbers of low (n = 4), medium (n = 11), and high-risk (n = 6) SIFK scores. One- and 2-year freedom-from-arthroplasty rates were both 100% for ISA subjects, and 76% and 55%, respectively, for Controls (P = 0.001 for cross-group comparison). Control knees with low, medium, and high-risk SIFK scores had respective 1- and 2-year survival rates of 100% and 100%, 90% and 68% (P = 0.07 vs. ISA), and 33% and 0% (P = 0.002 vs. ISA).

CONCLUSIONS

ISA intervention was strongly associated with avoidance of arthroplasty at a minimum 2 years, especially in patients with high-risk SIFK scores. SIFK severity scoring predicted relative risk of conversion to arthroplasty through at least 2 years in non-surgically treated subjects.

A guide to appropriately planning and conducting meta-analyses: part 3. Special considerations-the network meta-analysis

The meta-analysis has become one of the predominant studies designs in orthopaedic literature. Within recent years, the network meta-analysis has been implicated as a powerful approach to comparing multiple treatments for an outcome of interest when conducting a meta-analysis (as opposed to two competing treatments which is typical of a traditional meta-analysis). With the increasing use of the network meta-analysis, it is imperative for readers to possess the ability to independently and critically evaluate these types of studies. The purpose of this article is to provide the necessary foundation of knowledge to both properly conduct and interpret the results of a network meta-analysis.

Lower donor site morbidity with hamstring and quadriceps tendon autograft compared with bone-patellar tendon-bone autograft after anterior cruciate ligament reconstruction: a systematic review and network meta-analysis of randomized controlled trials

PURPOSE

To perform a meta-analysis of RCTs evaluating donor site morbidity after bone-patellar tendon-bone (BTB), hamstring tendon (HT) and quadriceps tendon (QT) autograft harvest for anterior cruciate ligament reconstruction (ACLR).

METHODS

PubMed, OVID/Medline and Cochrane databases were queried in July 2022. All level one articles reporting the frequency of specific donor-site morbidity were included. Frequentist model network meta-analyses with P-scores were conducted to compare the prevalence of donor-site morbidity, complications, all-cause reoperations and revision ACLR among the three treatment groups.

RESULTS

Twenty-one RCTs comprising the outcomes of 1726 patients were included. The overall pooled rate of donor-site morbidity (defined as anterior knee pain, difficulty/impossibility kneeling, or combination) was 47.3% (range, 3.8-86.7%). A 69% (95% confidence interval [95% CI]: 0.18-0.56) and 88% (95% CI: 0.04-0.33) lower odds of incurring donor-site morbidity was observed with HT and QT autografts, respectively (p < 0.0001, both), when compared to BTB autograft. QT autograft was associated with a non-statistically significant reduction in donor-site morbidity compared with HT autograft (OR: 0.37, 95% CI: 0.14-1.03, n.s.). Treatment rankings (ordered from best-to-worst autograft choice with respect to donor-site morbidity) were as follows: (1) QT (P-score = 0.99), (2) HT (P-score = 0.51) and (3) BTB (P-score = 0.00). No statistically significant associations were observed between autograft and complications (n.s.), reoperations (n.s.) or revision ACLR (n.s.).

CONCLUSION

ACLR using HT and QT autograft tissue was associated with a significant reduction in donor-site morbidity compared to BTB autograft. Autograft selection was not associated with complications, all-cause reoperations, or revision ACLR. Based on the current data, there is sufficient evidence to recommend that autograft selection should be personalized through considering differential rates of donor-site morbidity in the context of patient expectations and activity level without concern for a clinically important change in the rate of adverse events.

LEVEL OF EVIDENCE

Level I.

A guide to appropriately planning and conducting meta-analyses: part 2-effect size estimation, heterogeneity and analytic approaches

Meta-analyses by definition are a subtype of systematic review intended to quantitatively assess the strength of evidence present on an intervention or treatment. Such analyses may use individual-level data or aggregate data to produce a point estimate of an effect, also known as the combined effect, and measure precision of the calculated estimate. The current article will review several important considerations during the analytic phase of a meta-analysis, including selection of effect estimators, heterogeneity and various sub-types of meta-analytic approaches.

Survival analyses and their applications in orthopaedics

Survival analyses are a powerful statistical tool used to analyse data when the outcome of interest involves the time until an event. There is an array of models fit for this goal; however, there are subtle differences in assumptions, as well as a number of pitfalls, that can lead to biased results if researchers are unaware of the subtleties. As larger amounts of data become available, and more survival analyses are published every year, it is important that healthcare professionals understand how to evaluate these models and apply them into their practice. Therefore, the purpose of this study was to present an overview of survival analyses, including required assumptions and important pitfalls, as well as examples of their use within orthopaedic surgery.

Unsupervised machine learning methods and emerging applications in healthcare

Unsupervised machine learning methods are important analytical tools that can facilitate the analysis and interpretation of high-dimensional data. Unsupervised machine learning methods identify latent patterns and hidden structures in high-dimensional data and can help simplify complex datasets. This article provides an overview of key unsupervised machine learning techniques including K-means clustering, hierarchical clustering, principal component analysis, and factor analysis. With a deeper understanding of these analytical tools, unsupervised machine learning methods can be incorporated into health sciences research to identify novel risk factors, improve prevention strategies, and facilitate delivery of personalized therapies and targeted patient care.Level of evidence: I.

A practical guide to the development and deployment of deep learning models for the Orthopedic surgeon: part I

Deep learning has a profound impact on daily life. As Orthopedics makes use of this rapid escalation in technology, Orthopedic surgeons will need to take leadership roles on deep learning projects. Moreover, surgeons must possess an understanding of what is necessary to design and implement deep learning-based project pipelines. This review provides a practical guide for the Orthopedic surgeon to understand the steps needed to design, develop, and deploy a deep learning pipeline for clinical applications. A detailed description of the processes involved in defining the problem, building the team, acquiring and curating the data, labeling the data, establishing the ground truth, pre-processing and augmenting the data, and selecting the required hardware is provided. In addition, an overview of unique considerations involved in the training and evaluation of deep learning models is provided. This review strives to provide surgeons with the groundwork needed to identify gaps in the clinical landscape that deep learning models may be able to fill and equips them with the knowledge needed to lead an interdisciplinary team through the process of creating novel deep-learning-based solutions to fill those gaps.

Evaluating the Need for Preoperative MRI Before Primary Hip Arthroscopy in Patients 40 Years and Younger With Femoroacetabular Impingement Syndrome: A Multicenter Comparative Analysis

Background

Routine hip magnetic resonance imaging (MRI) before arthroscopy for patients with femoroacetabular impingement syndrome (FAIS) offers questionable clinical benefit, delays surgery, and wastes resources.

Purpose

To assess the clinical utility of preoperative hip MRI for patients aged ≤40 years who were undergoing primary hip arthroscopy and who had a history, physical examination findings, and radiographs concordant with FAIS.

Study Design

Cohort study; Level of evidence, 3.

Methods

Included were 1391 patients (mean age, 25.8 years; 63% female; mean body mass index, 25.6) who underwent hip arthroscopy between August 2015 and December 2021 by 1 of 4 fellowship-trained hip surgeons from 4 referral centers. Inclusion criteria were FAIS, primary surgery, and age ≤40 years. Exclusion criteria were MRI contraindication, reattempt of nonoperative management, and concomitant periacetabular osteotomy. Patients were stratified into those who were evaluated with preoperative MRI versus those without MRI. Those without MRI received an MRI before surgery without deviation from the established surgical plan. All preoperative MRI scans were compared with the office evaluation and intraoperative findings to assess agreement. Time from office to arthroscopy and/or MRI was recorded. MRI costs were calculated.

Results

Of the study patients, 322 were not evaluated with MRI and 1069 were. MRI did not alter surgical or interoperative plans. Both groups had MRI findings demonstrating anterosuperior labral tears treated intraoperatively (99.8% repair, 0.2% debridement, and 0% reconstruction). Compared with patients who were evaluated with MRI and waited 63.0 ± 34.6 days, patients who were not evaluated with MRI underwent surgery 6.5 ± 18.7 days after preoperative MRI. MRI delayed surgery by 24.0 ± 5.3 days and cost a mean $2262 per patient.

Conclusion

Preoperative MRI did not alter indications for primary hip arthroscopy in patients aged ≤40 years with a history, physical examination findings, and radiographs concordant with FAIS. Rather, MRI delayed surgery and wasted resources. Routine hip MRI acquisition for the younger population with primary FAIS with a typical presentation should be challenged.

Choosing the appropriate measure of central tendency: mean, median, or mode?

Mean, median, and mode are among the most basic and consistently used measures of central tendency in statistical analysis and are crucial for simplifying data sets to a single value. However, there is a lack of understanding of when to use each metric and how various factors can impact these values. The aim of this article is to clarify some of the confusion related to each measure and explain how to select the appropriate metric for a given data set. The authors present this work as an educational resource, ensuring that these common statistical concepts are better understood throughout the Orthopedic research community.

Multivariable regression: understanding one of medicine's most fundamental statistical tools

Multivariable regression is a fundamental tool that drives observational research in orthopaedic surgery. However, regression analyses are not always implemented correctly. This study presents a basic overview of regression analyses and reviews frequent points of confusion. Topics include linear, logistic, and time-to-event regressions, causal inference, confounders, overfitting, missing data, multicollinearity, interactions, and key differences between multivariable versus multivariate regression. The goal is to provide clarity regarding the use and interpretation of multivariable analyses for those attempting to increase their statistical literacy in orthopaedic research.

Natural language processing: using artificial intelligence to understand human language in orthopedics

Natural language processing (NLP) describes the broad field of artificial intelligence by which computers are trained to understand and generate human language. Within healthcare research, NLP is commonly used for variable extraction and classification/cohort identification tasks. While these tools are becoming increasingly popular and available as both open-source and commercial products, there is a paucity of the literature within the orthopedic space describing the key tasks within these powerful pipelines. Curation and navigation of the electronic medical record are becoming increasingly onerous, and it is important for physicians and other healthcare professionals to understand potential methods of harnessing this large data resource. The purpose of this study is to provide an overview of the tasks required to develop an NLP pipeline for orthopedic research and present recent examples of successful implementations.

Comparison of Posterior Cruciate Ligament Reconstruction Using an All-Inside Technique With and Without Independent Suture Tape Reinforcement

Background

Biomechanical studies support the use of suture tape reinforcement for limiting graft elongation and increasing strength in knee ligament reconstructions.

Purpose

To compare posterior cruciate ligament (PCL) laxity, complication and reoperation rates, and patient-reported outcomes (PROs) after all-inside single-bundle PCL reconstruction (PCLR) with versus without independent suture tape reinforcement.

Study Design

Cohort study; Level of evidence, 3.

Methods

A retrospective cohort study of consecutive patients who underwent primary, all-inside allograft single-bundle PCLR with and without independent suture tape reinforcement at a single academic institution from 2012 to 2019. Medical records were reviewed for patient characteristics, additional injuries, and concomitant procedures. PRO scores (including the International Knee Documentation Committee [IKDC], Tegner activity scale, and Lysholm scores), bilateral comparison kneeling radiographs, and physical examination findings were collected at a minimum of 2 years postoperatively.

Results

Included were 50 patients: 19 with suture tape reinforcement (mean age 30.6 ± 2.9 years) and 31 without suture tape reinforcement (control group; mean age 26.2 ± 1.6 years). One PCLR graft in the suture tape group failed. Posterior drawer examination revealed grade 1+ laxity in 4 of 19 (21%) of the suture tape cohort versus 6 of 31 (19%) of the control cohort (P > .999). Bilateral kneeling radiographs showed similar side-to-side differences in laxity between the groups (suture tape vs control: mean, 1.9 ± 0.4 vs 2.6 ± 0.6 mm; P = .361). There were no statistically significant differences between the groups in postoperative IKDC (suture tape vs control: 79.3 vs 79.6; P = .779), Lysholm (87.5 vs 84.3; P = .828), or Tegner activity (5.6 vs 5.7; P = .562) scores.

Conclusion

All-inside single-bundle PCLR with and without independent suture tape reinforcement demonstrated low rates of graft failure, complications, and reoperations, with satisfactory PROs at a minimum 2-year follow-up. Radiographic posterior tibial translation was comparable between the 2 groups.

Supervised machine learning and associated algorithms: applications in orthopedic surgery

Supervised learning is the most common form of machine learning utilized in medical research. It is used to predict outcomes of interest or classify positive and/or negative cases with a known ground truth. Supervised learning describes a spectrum of techniques, ranging from traditional regression modeling to more complex tree boosting, which are becoming increasingly prevalent as the focus on "big data" develops. While these tools are becoming increasingly popular and powerful, there is a paucity of literature available that describe the strengths and limitations of these different modeling techniques. Typically, there is no formal training for health care professionals in the use of machine learning models. As machine learning applications throughout medicine increase, it is important that physicians and other health care professionals better understand the processes underlying application of these techniques. The purpose of this study is to provide an overview of commonly used supervised learning techniques with recent case examples within the orthopedic literature. An additional goal is to address disparities in the understanding of these methods to improve communication within and between research teams.

Limited clinical utility of a machine learning revision prediction model based on a national hip arthroscopy registry

PURPOSE

Accurate prediction of outcome following hip arthroscopy is challenging and machine learning has the potential to improve our predictive capability. The purpose of this study was to determine if machine learning analysis of the Danish Hip Arthroscopy Registry (DHAR) can develop a clinically meaningful calculator for predicting the probability of a patient undergoing subsequent revision surgery following primary hip arthroscopy.

METHODS

Machine learning analysis was performed on the DHAR. The primary outcome for the models was probability of revision hip arthroscopy within 1, 2, and/or 5 years after primary hip arthroscopy. Data were split randomly into training (75%) and test (25%) sets. Four models intended for these types of data were tested: Cox elastic net, random survival forest, gradient boosted regression (GBM), and super learner. These four models represent a range of approaches to statistical details like variable selection and model complexity. Model performance was assessed by calculating calibration and area under the curve (AUC). Analysis was performed using only variables available in the pre-operative clinical setting and then repeated to compare model performance using all variables available in the registry.

RESULTS

In total, 5581 patients were included for analysis. Average follow-up time or time-to-revision was 4.25 years (± 2.51) years and overall revision rate was 11%. All four models were generally well calibrated and demonstrated concordance in the moderate range when restricted to only pre-operative variables (0.62-0.67), and when considering all variables available in the registry (0.63-0.66). The 95% confidence intervals for model concordance were wide for both analyses, ranging from a low of 0.53 to a high of 0.75, indicating uncertainty about the true accuracy of the models.

CONCLUSION

The association between pre-surgical factors and outcome following hip arthroscopy is complex. Machine learning analysis of the DHAR produced a model capable of predicting revision surgery risk following primary hip arthroscopy that demonstrated moderate accuracy but likely limited clinical usefulness. Prediction accuracy would benefit from enhanced data quality within the registry and this preliminary study holds promise for future model generation as the DHAR matures. Ongoing collection of high-quality data by the DHAR should enable improved patient-specific outcome prediction that is generalisable across the population.

LEVEL OF EVIDENCE

Level III.

Machine Learning for Predicting Lower Extremity Muscle Strain in National Basketball Association Athletes

Background

In professional sports, injuries resulting in loss of playing time have serious implications for both the athlete and the organization. Efforts to quantify injury probability utilizing machine learning have been met with renewed interest, and the development of effective models has the potential to supplement the decision-making process of team physicians.

Purpose/Hypothesis

The purpose of this study was to (1) characterize the epidemiology of time-loss lower extremity muscle strains (LEMSs) in the National Basketball Association (NBA) from 1999 to 2019 and (2) determine the validity of a machine-learning model in predicting injury risk. It was hypothesized that time-loss LEMSs would be infrequent in this cohort and that a machine-learning model would outperform conventional methods in the prediction of injury risk.

Study Design

Case-control study; Level of evidence, 3.

Methods

Performance data and rates of the 4 major muscle strain injury types (hamstring, quadriceps, calf, and groin) were compiled from the 1999 to 2019 NBA seasons. Injuries included all publicly reported injuries that resulted in lost playing time. Models to predict the occurrence of a LEMS were generated using random forest, extreme gradient boosting (XGBoost), neural network, support vector machines, elastic net penalized logistic regression, and generalized logistic regression. Performance was compared utilizing discrimination, calibration, decision curve analysis, and the Brier score.

Results

A total of 736 LEMSs resulting in lost playing time occurred among 2103 athletes. Important variables for predicting LEMS included previous number of lower extremity injuries; age; recent history of injuries to the ankle, hamstring, or groin; and recent history of concussion as well as 3-point attempt rate and free throw attempt rate. The XGBoost machine achieved the best performance based on discrimination assessed via internal validation (area under the receiver operating characteristic curve, 0.840), calibration, and decision curve analysis.

Conclusion

Machine learning algorithms such as XGBoost outperformed logistic regression in the prediction of a LEMS that will result in lost time. Several variables increased the risk of LEMS, including a history of various lower extremity injuries, recent concussion, and total number of previous injuries.

Duration of Care and Operative Time Are the Primary Drivers of Total Charges After Ambulatory Hip Arthroscopy: A Machine Learning Analysis

PURPOSE

To develop a machine learning algorithm to predict total charges after ambulatory hip arthroscopy and create a risk-adjusted payment model based on patient comorbidities.

METHODS

A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective hip arthroscopy between 2015 and 2016. Features included in initial models consisted of patient characteristics, medical comorbidities, and procedure-specific variables. Models were generated to predict total charges using 5 algorithms. Model performance was assessed by the root-mean-square error, root-mean-square logarithmic error, and coefficient of determination. Global variable importance and partial dependence curves were constructed to show the impact of each input feature on total charges. For performance benchmarking, the best candidate model was compared with a multivariate linear regression using the same input features.

RESULTS

A total of 5,121 patients were included. The median cost after hip arthroscopy was $19,720 (interquartile range, $12,399-$26,439). The gradient-boosted ensemble model showed the best performance (root-mean-square error, $3,800 [95% confidence interval, $3,700-$3,900]; logarithmic root-mean-square error, 0.249 [95% confidence interval, 0.24-0.26]; R² = 0.73). Major cost drivers included total hours in facility less than 12 or more than 15, longer procedure time, performance of a labral repair, age younger than 30 years, Elixhauser Comorbidity Index (ECI) of 1 or greater, African American race, residence in extreme urban and rural areas, and higher household and neighborhood income.

CONCLUSIONS

The gradient-boosted ensemble model effectively predicted total charges after hip arthroscopy. Few modifiable variables were identified other than anesthesia type; nonmodifiable drivers of total charges included duration of care less than 12 hours or more than 15 hours, operating room time more than 100 minutes, age younger than 30 years, performance of a labral repair, and ECI greater than 0. Stratification of patients based on the ECI highlighted the increased financial risk borne by physicians via flat reimbursement schedules given variable degrees of comorbidities.

LEVEL OF EVIDENCE

Level III, retrospective cohort study.

A Machine Learning Algorithm Outperforms Traditional Multiple Regression to Predict Risk of Unplanned Overnight Stay Following Outpatient Medial Patellofemoral Ligament Reconstruction

Purpose

To determine whether conventional logistic regression or machine learning algorithms were more precise in identifying the risk factors for unplanned overnight admission after medial patellofemoral ligament (MPFL) reconstruction.

Methods

A retrospective review of the prospectively collected National Surgical Quality Improvement Program database was performed to identify patients who underwent outpatient MPFL reconstruction from 2006–2018. Patients admitted overnight were identified as those with length of stay of 1 or more days. Models were generated using random forest, extreme gradient boosting, adaptive boosting, or elastic net penalized logistic regression, and an additional model was produced as a weighted ensemble of the 4 final algorithms. The predictive capacity of these models was compared to that of logistic regression.

Results

Of the 1307 patients identified, 221 (16.9%) required at least one overnight stay after MPFL reconstruction. Multivariate logistic regression found the following variables to be predictors of inpatient admission: age (odds ratio [OR] = 1.03 [95% confidence interval {CI} 1.02-1.04]; P <.001), spinal anesthesia (OR = 3.42 [95% CI 1.98-6.08]; P < .001), American Society of Anesthesiologists (ASA) class 3/4 (OR = 1.96 [95% CI 1.25-3.06]; P < .001), history of chronic obstructive pulmonary disease (COPD) (OR = 6.44 [95% CI 1.58-26.17]; P = .02), and body mass index (BMI) (OR = 1.03 [95% CI 1.01-1.05]; P < .001). The ensemble model achieved the best performance based on discrimination assessed via internal validation (area under the curve = 0.722). The variables determined most important by the ensemble model were increasing BMI, increasing age, ASA class, anesthesia, smoking, hypertension, lateral release, and history of COPD.

Conclusions

An internally validated machine learning algorithm outperformed logistic regression modeling in predicting the need for unplanned overnight hospitalization after MPFL reconstruction. In this model, the most significant risk factors for admission were age, BMI, ASA class, smoking status, hypertension, lateral release, and history of COPD. This tool can be deployed to augment provider assessment to identify high-risk candidates and appropriately set postoperative expectations for patients.

Clinical Relevance

Identifying and mitigating patient risk factors to prevent adverse surgical outcomes and hospitalizations is one of our primary goals. There may be a key role for machine learning algorithms to help successfully and efficiently risk stratify patients to decrease costs, appropriately set postoperative expectations, and increase the quality of delivered care.

Editorial Commentary: Machine Learning in Medicine Requires Clinician Input, Faces Barriers, and High-Quality Evidence Is Required to Demonstrate Improved Patient Outcomes

Machine learning (ML) and artificial intelligence (AI) may be described as advanced statistical techniques using algorithms to "learn" to evaluate and predict relationships between input and results without explicit human programming, often with high accuracy. The potentials and pitfalls of ML continue to be explored as predictive modeling grows in popularity. While use of and optimism for AI continues to increase in orthopaedic surgery, there remains little high-quality evidence of its ability to improve patient outcome. It is up to us as clinicians to provide context for ML models and guide the use of these technologies to optimize the outcome for our patients. Barriers to widespread adoption of ML include poor quality data, limits to compliant data sharing, few clinicians who are expert in ML statistical techniques, and computing costs including technology, infrastructure, personnel, energy, and updates.

Neurological Complications Following Arthroscopic and Related Sports Surgery: Prevention, Work-up, and Treatment

Arthroscopy of the shoulder, elbow, hip, and knee has become increasingly utilized due to continued advancements in technique, training, and instrumentation. In addition, arthroscopy is generally safe and effective in the utilization of joint preservation surgical techniques. The arthroscopist must utilize a thorough understanding of the surgical anatomy, detailed care with patient positioning, and safe instrumentation portals to prevent associated neurological injury. In the event of postoperative neurological complications, the physician must carefully document the patient history and physical examination while considering the utilization of additional imaging, testing, or surgical nerve exploration with a specialized team depending upon the severity of neurological injury. In this review, we discuss the prevention, evaluation, and treatment of neurological complications related for arthroscopic procedures of the shoulder, elbow, hip, and knee.

Machine learning and conventional statistics: making sense of the differences

The application of machine learning (ML) to the field of orthopaedic surgery is rapidly increasing, but many surgeons remain unfamiliar with the nuances of this novel technique. With this editorial, we address a fundamental topic-the differences between ML techniques and traditional statistics. By doing so, we aim to further familiarize the reader with the new opportunities available thanks to the ML approach.

Artificial intelligence and machine learning: an introduction for orthopaedic surgeons

The application of artificial intelligence (AI) and machine learning to the field of orthopaedic surgery is rapidly increasing. While this represents an important step in the advancement of our specialty, the concept of AI is rich with statistical jargon and techniques unfamiliar to many clinicians. This knowledge gap may limit the impact and potential of these novel techniques. We aim to narrow this gap in a way that is accessible for all orthopaedic surgeons. With this manuscript, we introduce the concept of AI and machine learning and give examples of how it can impact clinical practice and patient care.Level of evidence VI.

Predicting Anterior Cruciate Ligament Reconstruction Revision: A Machine Learning Analysis Utilizing the Norwegian Knee Ligament Register

BACKGROUND

Several factors are associated with an increased risk of anterior cruciate ligament (ACL) reconstruction revision. However, the ability to accurately translate these factors into a quantifiable risk of revision at a patient-specific level has remained elusive. We sought to determine if machine learning analysis of the Norwegian Knee Ligament Register (NKLR) can identify the most important risk factors associated with subsequent revision of primary ACL reconstruction and develop a clinically meaningful calculator for predicting revision of primary ACL reconstruction.

METHODS

Machine learning analysis was performed on the NKLR data set. The primary outcome was the probability of revision ACL reconstruction within 1, 2, and/or 5 years. Data were split randomly into training sets (75%) and test sets (25%). Four machine learning models were tested: Cox Lasso, survival random forest, generalized additive model, and gradient boosted regression. Concordance and calibration were calculated for all 4 models.

RESULTS

The data set included 24,935 patients, and 4.9% underwent a revision surgical procedure during a mean follow-up (and standard deviation) of 8.1 ± 4.1 years. All 4 models were well-calibrated, with moderate concordance (0.67 to 0.69). The Cox Lasso model required only 5 variables for outcome prediction. The other models either used more variables without an appreciable improvement in accuracy or had slightly lower accuracy overall. An in-clinic calculator was developed that can estimate the risk of ACL revision (Revision Risk Calculator). This calculator can quantify risk at a patient-specific level, with a plausible range from near 0% for low-risk patients to 20% for high-risk patients at 5 years.

CONCLUSIONS

Machine learning analysis of a national knee ligament registry can predict the risk of ACL reconstruction revision with moderate accuracy. This algorithm supports the creation of an in-clinic calculator for point-of-care risk stratification based on the input of only 5 variables. Similar analysis using a larger or more comprehensive data set may improve the accuracy of risk prediction, and future studies incorporating patients who have experienced failure of ACL reconstruction but have not undergone subsequent revision may better predict the true risk of ACL reconstruction failure.

LEVEL OF EVIDENCE

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Racial and Gender Shoulder Arthroplasty Utilization Disparities of High- and Low-Volume Centers in New York State

Introduction

The literature has consistently demonstrated utilization disparities in joint replacement procedures, though no studies have evaluated disparities in total shoulder arthroplasty with regard to operative volume.

Methods

We queried the New York (NY) Statewide Planning and Research Cooperative System (SPARCS) database for 32 410 total shoulder arthroplasties performed between 2009 and 2017. Patients were identified using Clinical Classifications Software code 154 for Non-Hip/Knee Arthroplasty and All Patient Refined-Diagnosis Related Group code 322 for Shoulder. Racial groups included Hispanic, non-Hispanic white, non-Hispanic black, and Other. High-volume centers were facilities that performed 2 standard deviations above the mean annual procedures. Utilization rates were calculated by dividing total shoulder arthroplasties per group by the 2010 NY Census population of that group. The Fisher exact test was used to determine significance.

Results

Total shoulder arthroplasty utilization increased from 43/100 000 to 73/100 000, two-thirds of which was driven by an increase in white resident utilization. More White residents per 100 000 underwent shoulder arthroplasty than Black, Hispanic, and Other residents per 100 000 residents of their respective race. White residents were 90% more likely than Hispanic residents to undergo total shoulder arthroplasty at high-volume centers (P = .04). There were no differences in utilization rate regarding operative volume comparing Black or Other residents to White residents. More females underwent total shoulder arthroplasty than males, though there was no difference in utilization rate regarding operative volume.

Conclusion

Though total shoulder arthroplasty utilization nearly doubled, disparities persisted across gender and minority groups particularly in Hispanic utilization as White residents were 90% more likely than Hispanic residents to undergo shoulder arthroplasty at high-volume centers.

Machine learning algorithm to predict anterior cruciate ligament revision demonstrates external validity

PURPOSE

External validation of machine learning predictive models is achieved through evaluation of model performance on different groups of patients than were used for algorithm development. This important step is uncommonly performed, inhibiting clinical translation of newly developed models. Machine learning analysis of the Norwegian Knee Ligament Register (NKLR) recently led to the development of a tool capable of estimating the risk of anterior cruciate ligament (ACL) revision ( https://swastvedt.shinyapps.io/calculator_rev/ ). The purpose of this study was to determine the external validity of the NKLR model by assessing algorithm performance when applied to patients from the Danish Knee Ligament Registry (DKLR).

METHODS

The primary outcome measure of the NKLR model was probability of revision ACL reconstruction within 1, 2, and/or 5 years. For external validation, all DKLR patients with complete data for the five variables required for NKLR prediction were included. The five variables included graft choice, femur fixation device, KOOS QOL score at surgery, years from injury to surgery, and age at surgery. Predicted revision probabilities were calculated for all DKLR patients. The model performance was assessed using the same metrics as the NKLR study: concordance and calibration.

RESULTS

In total, 10,922 DKLR patients were included for analysis. Average follow-up time or time-to-revision was 8.4 (± 4.3) years and overall revision rate was 6.9%. Surgical technique trends (i.e., graft choice and fixation devices) and injury characteristics (i.e., concomitant meniscus and cartilage pathology) were dissimilar between registries. The model produced similar concordance when applied to the DKLR population compared to the original NKLR test data (DKLR: 0.68; NKLR: 0.68-0.69). Calibration was poorer for the DKLR population at one and five years post primary surgery but similar to the NKLR at two years.

CONCLUSION

The NKLR machine learning algorithm demonstrated similar performance when applied to patients from the DKLR, suggesting that it is valid for application outside of the initial patient population. This represents the first machine learning model for predicting revision ACL reconstruction that has been externally validated. Clinicians can use this in-clinic calculator to estimate revision risk at a patient specific level when discussing outcome expectations pre-operatively. While encouraging, it should be noted that the performance of the model on patients undergoing ACL reconstruction outside of Scandinavia remains unknown.

LEVEL OF EVIDENCE

III.

Understanding Anterior Shoulder Instability Through Machine Learning: New Models That Predict Recurrence, Progression to Surgery, and Development of Arthritis

Background

Management of anterior shoulder instability (ASI) aims to reduce risk of future recurrence and prevent complications via nonoperative and surgical management. Machine learning may be able to reliably provide predictions to improve decision making for this condition.

Purpose

To develop and internally validate a machine-learning model to predict the following outcomes after ASI: (1) recurrent instability, (2) progression to surgery, and (3) the development of symptomatic osteoarthritis (OA) over long-term follow-up.

Study Design

Cohort study (prognosis); Level of evidence, 2.

Methods

An established geographic database of >500,000 patients was used to identify 654 patients aged <40 years with an initial diagnosis of ASI between 1994 and 2016; the mean follow-up was 11.1 years. Medical records were reviewed to obtain patient information, and models were generated to predict the outcomes of interest. Five candidate algorithms were trained in the development of each of the models, as well as an additional ensemble of the algorithms. Performance of the algorithms was assessed using discrimination, calibration, and decision curve analysis.

Results

Of the 654 included patients, 443 (67.7%) experienced multiple instability events, 228 (34.9%) underwent surgery, and 39 (5.9%) developed symptomatic OA. The ensemble gradient-boosted machines achieved the best performances based on discrimination (via area under the receiver operating characteristic curve [AUC]: AUC_recurrence = 0.86), AUC_surgery = 0.76, AUC_OA = 0.78), calibration, decision curve analysis, and Brier score (Brier_recurrence = 0.138, Brier_surgery = 0.185, Brier_OA = 0.05). For demonstration purposes, models were integrated into a single web-based open-access application able to provide predictions and explanations for practitioners and researchers.

Conclusion

After identification of key features, including time from initial instability, age at initial instability, sports involvement, and radiographic findings, machine-learning models were developed that effectively and reliably predicted recurrent instability, progression to surgery, and the development of OA in patients with ASI. After careful external validation, these models can be incorporated into open-access digital applications to inform patients, clinicians, and researchers regarding quantifiable risks of relevant outcomes in the clinic.

Are Baseball Statistics an Appropriate Tool for Assessing Return to Play in Injured Pitchers? Analysis of Statistical Variability in Healthy Players

Background:

Basic pitcher statistics have been used to assess performance in pitchers after injury or surgery without being validated. Even among healthy pitchers, the normal variability of these parameters has not yet been established.

Purpose:

To determine (1) the normal variability of basic and advanced pitcher statistics in healthy professional baseball pitchers and (2) the minimum pitches needed to predict these parameters.

Study Design:

Cross-sectional study; Level of evidence, 3.

Methods:

Publicly available data from the MLB Statcast and PITCHf/x databases were used to analyze MLB pitchers during the 2015 and 2016 seasons who recorded a minimum of 100 innings without injury. Basic and advanced baseball pitcher statistics were analyzed. The variability of each parameter was assessed by computing the coefficient of variation (CV) between individual pitchers and across all pitchers. A CV <10 was indicative of a relatively constant parameter, and parameters with a CV >10 were generally considered inconsistent and unreliable. The minimum number of pitches needed to be followed for each variable was also analyzed.

Results:

A total of 118 pitchers, 55 baseball-specific statistical metrics (38 basic and 17 advanced), and 7.5 million pitches were included and analyzed. Of the 38 basic pitcher statistics, only fastball velocity demonstrated a CV <10 (CV = 1.5), while 6 of 17 (35%) advanced metrics demonstrated acceptable consistency (CV <10). Release position from plate and velocity from the plate were the 2 most consistent advanced parameters. When separated by pitch type, these 2 parameters were the most constant (lowest CV) across every pitch type.

Conclusion:

We recommend against utilizing nonvalidated statistical measures to assess performance after injury, as they demonstrated unacceptably high variability even among healthy, noninjured professional baseball pitchers. It is our hope that this study will serve as the foundation for the identification and implementation of validated pitcher-dependent statistical measures that can be used to assess return-to-play performance after injury in the future.

Automated Risk Stratification of Hip Osteoarthritis Development in Patients With Femoroacetabular Impingement Using an Unsupervised Clustering Algorithm: A Study From the Rochester Epidemiology Project

Background:

Studies evaluating the natural history of femoroacetabular impingement (FAI) are limited.

Purpose:

To stratify the risk of progression to osteoarthritis (OA) in patients with FAI using an unsupervised machine-learning algorithm, compare the characteristics of each subgroup, and validate the reproducibility of staging.

Study Design:

Cohort study (prognosis); Level of evidence, 2.

Methods:

A geographic database from the Rochester Epidemiology Project was used to identify patients with hip pain between 2000 and 2016. Medical charts were reviewed to obtain characteristic information, physical examination findings, and imaging details. The patient data were randomly split into 2 mutually exclusive sets: train set (70%) for model development and test set (30%) for validation. The data were transformed via Uniform Manifold Approximation and Projection and were clustered using Hierarchical Density-based Spatial Clustering of Applications with Noise.

Results:

The study included 1071 patients with a mean follow-up period of 24.7 ± 12.5 years. The patients were clustered into 5 subgroups based on train set results: patients in cluster 1 were in their early 20s (20.9 ± 9.6 years), female dominant (84%), with low body mass index (<19 ); patients in cluster 2 were in their early 20s (22.9 ± 6.7 years), female dominant (95%), and pincer-type FAI (100%) dominant; patients in cluster 3 were in their mid 20s (26.4 ± 9.7) and were mixed-type FAI dominant (92%); patients in cluster 4 were in their early 30s (32.7 ± 7.8), with high body mass index (≥29 ), and diabetes (17%); and patients in cluster 5 were in their early 30s (30.0 ± 9.1), with a higher percentage of males (43%) compared with the other clusters and with limited internal rotation (14%). Mean survival for clusters 1 to 5 was 17.9 ± 0.6, 18.7 ± 0.3, 17.1 ± 0.4, 15.0 ± 0.5, and 15.6 ± 0.5 years, respectively, in the train set. The survival difference was significant between clusters 1 and 4 (P = .02), 2 and 4 (P < .005), 2 and 5 (P = .01), and 3 and 4 (P < .005) in the train set and between clusters 2 and 5 (P = .03) and 3 and 4 (P = .01) in the test set. Cluster characteristics and prognosis was well reproduced in the test set.

Conclusion:

Using the clustering algorithm, it was possible to determine the prognosis for OA progression in patients with FAI in the presence of conflicting risk factors acting in combination.

Post-arthroscopic Subchondral Insufficiency Fractures of the Knee Yield High Rate of Conversion to Arthroplasty

PURPOSE

To evaluate both the potential causes and resultant outcomes in patients in whom subchondral insufficiency fracture of the knee (SIFK) develops after arthroscopy.

METHODS

We performed a retrospective review of all patients with a magnetic resonance imaging diagnosis of SIFK after arthroscopic meniscectomy and chondroplasty over a 12-year period.

RESULTS

A total of 28 patients were included, with a mean age of 61 years and mean follow-up period of 5.7 years. SIFK showed a predilection for the medial compartment (n = 25, 89%), specifically the medial femoral condyle (n = 21, 75%). In 7 patients (25%), SIFK developed in both the femoral condyle and tibial plateau in the ipsilateral compartment. Fifteen patients (54%) went on to conversion to arthroplasty at a mean of 0.72 years. The rate of survival free of conversion to arthroplasty was 57%, 45%, and 40% at 1 year, 2 years, and 5 years, respectively. Furthermore, 63% of patients with a meniscal tear and SIFK in the same compartment went on to arthroplasty (P = .04). There was an increased risk of conversion to arthroplasty if SIFK was present in both the femur and tibia in the same compartment (P = .04). A higher Kellgren-Lawrence grade at the time of the SIFK diagnosis increased the likelihood of eventual arthroplasty (P = .03). The presence of SIFK in both the femur and tibia in the ipsilateral compartment, an increased Kellgren-Lawrence grade, and a meniscal tear or prior meniscectomy in the same compartment as SIFK were associated with an increased risk of eventual arthroplasty.

CONCLUSIONS

Post-arthroscopic SIFK most commonly occurs in the medial compartment, particularly in patients who underwent a prior meniscectomy. The presence of meniscal root and radial tears in these patients is notable (75%). Ultimately, there is a high rate of progression of arthrosis (33%) and eventual conversion to arthroplasty.

LEVEL OF EVIDENCE

Level IV, case series.

All-Inside PCL Reconstruction

Many different techniques with multiple graft types have been described for the reconstruction of the injured posterior cruciate ligament (PCL); autograft versus allograft, single- versus double-bundle, open inlay versus arthroscopic inlay versus arthroscopic transtibial, and recently described the arthroscopic "all-inside" socket technique. Reported clinical outcomes have demonstrated no significant difference in any of these PCL reconstruction techniques, likely because of the heterogeneity in injury characteristics and patient population. The ideal surgical technique should be safe, simple, and reproducible while allowing treatment of concomitant knee injuries resulting and return to function.

Clinical Characteristics and Outcomes in Patients with COVID-19 and Cancer: a Systematic Review and Meta-analysis

Much of routine cancer care has been disrupted due to the perceived susceptibility to SARS-CoV-2 infection in cancer patients. Here, we systematically review the current evidence base pertaining to the prevalence, presentation and outcome of COVID-19 in cancer patients, in order to inform policy and practice going forwards. A keyword-structured systematic search was conducted on Pubmed, Cochrane, Embase and MedRxiv databases for studies reporting primary data on COVID-19 in cancer patients. Studies were critically appraised using the NIH National Heart, Lung and Blood Institute's quality assessment tool set. The pooled prevalence of cancer as a co-morbidity in patients with COVID-19 and pooled in-hospital mortality risk of COVID-19 in cancer patients were derived by random-effects meta-analyses. In total, 110 studies from 10 countries were included. The pooled prevalence of cancer as a co-morbidity in hospitalised patients with COVID-19 was 2.6% (95% confidence interval 1.8%, 3.5%, I2: 92.0%). Specifically, 1.7% (95% confidence interval 1.3%, 2.3%, I2: 57.6.%) in China and 5.6% (95% confidence interval 4.5%, 6.7%, I2: 82.3%) in Western countries. Patients most commonly presented with non-specific symptoms of fever, dyspnoea and chest tightness in addition to decreased arterial oxygen saturation, ground glass opacities on computer tomography and non-specific changes in inflammatory markers. The pooled in-hospital mortality risk among patients with COVID-19 and cancer was 14.1% (95% confidence interval 9.1%, 19.8%, I2: 52.3%). We identified impeding questions that need to be answered to provide the foundation for an iterative review of the developing evidence base, and inform policy and practice going forwards. Analyses of the available data corroborate an unfavourable outcome of hospitalised patients with COVID-19 and cancer. Our findings encourage future studies to report detailed social, demographic and clinical characteristics of cancer patients, including performance status, primary cancer type and stage, as well as a history of anti-cancer therapeutic interventions.