Lu, Yining

Objectives: An increased posterior tibial slope (PTS) corresponds with an increased risk of graft failure following anterior cruciate ligament reconstruction (ACLR). Validated methods of human PTS measurements are subject to interobserver variability and can be inefficient on large-scale sets of images. The purpose of this study is to develop a deep learning artificial intelligence technique for the automated measurement of PTS from standard lateral knee radiographs. Methods: A deep learning U-Net model was developed on a cohort of 300 postoperative short leg lateral radiographs from ACLR patients to segment the tibial shaft, tibial joint surface, and tibial tuberosity. The model was trained via a random split following an 80:20 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 to human measurements performed by two study personnel using a previously validated manual technique for measuring PTS on short leg lateral radiographs on an independent test-set consisting of both preoperative 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 human-made and computer-vision measurements was 1.92⁰ (σ = 2.81⁰, P=0.24). Extreme disagreements between human and machine measurements as defined by differences ≥5⁰ occurred less than 5% of the time. The model was incorporated into a web based digital application for demonstration purposes which can provide measurement of an uploaded image in portable network graphics format in less than 5 seconds. Conclusions: We developed an accurate and reliable deep learning computer vision algorithm to automate the measurement of PTS on lateral knee radiographs. This tool will be deployed for clinical use on an institution-wide basis and, pending external validation, be made available to outside orthopaedic surgeons. This will provide an efficient and accurate tool to measure PTS during the preoperative assessment of ACL-injured patients.

Objectives: Desirable outcomes following surgery for anterior shoulder instability (ASI) include a multitude of functional and clinical functions, yet it remains uncertain if all of these outcomes are simultaneously achievable (i.e. can patients have full motion and no instability; does excellent stability coincide with stiffness?). The purpose of this study was to employ unsupervised machine learning techniques to define the actual "optimal observed outcome” for patients undergoing surgical treatment for ASI and to identify predictors of obtaining this outcome. Methods: Patients <40 years with an initial diagnosis of ASI from 1994 -2016 were included. Four unsupervised machine learning clustering algorithms were evaluated to partition subjects into “optimal observed outcome” or “suboptimal outcome” based on combinations of actually observed outcomes. Demographic, clinical, and treatment variables were compared between groups using descriptive statistics and Kaplan-Meier survival curves; multivariate stepwise logistic regression evaluated variable prognostic value. Results: 200 patients with a mean follow-up of 11 years were included. 146 (64%) obtained the “optimal observed outcome”, characterized by significantly (P <0.001) lower rates of: recurrent postoperative pain (23% vs 52%), recurrent instability (12% vs 41%), revision surgery (10% vs 24%), progression to osteoarthritis (OA) (5% vs. 19%), and mildly restricted motion (161° vs. 168°). Stepwise multivariate logistic regression identified time from initial instability to presentation (OR: 0.96, 95% CI: 0.92-0.98) and habitual instability (OR: 0.17, 95%CI: 0.04-0.77) as negative predictors of “observed optimal outcome”. Increased rate of subluxation over frank dislocation pre-operatively (OR: 1.30, 95% CI: 1.02-1.65) was a positive predictor. Type of surgery performed was not a significant predictor. Conclusions: Following surgical treatment for ASI, an appropriate “optimal observed outcome” can be defined as: minimal postoperative pain, absence of recurrent instability, low rates of revision surgery, absence of OA, and increased ROM. This “optimal observed outcome” was achieved in over 2/3rds of the cohort and this work demonstrated the synergistic relationship of these. The most significant predictors included shorter time to presentation and history of subluxations over frank dislocations pre-operatively. This definition of the “optimal observed outcome” may be more appropriate for surgical decision making and setting appropriate expectations. [Table: see text][Table: see text]

Purpose To use unsupervised machine learning clustering to define the “optimal observed outcome” after surgery for anterior shoulder instability (ASI) and to identify predictors for achieving it. Methods Medical records, images, and operative reports were reviewed for patients <40 years old undergoing surgery for ASI. Four unsupervised machine learning clustering algorithms partitioned subjects into “optimal observed outcome” or “suboptimal outcome” based on combinations of actually observed outcomes. Demographic, clinical, and treatment variables were compared between groups using descriptive statistics and Kaplan‐Meier survival curves. Variables were assessed for prognostic value through multivariate stepwise logistic regression. Results Two hundred patients with a mean follow‐up of 11 years were included. Of these, 146 (64%) obtained the “optimal observed outcome,” characterized by decreased: postoperative pain (23% vs 52%; P < 0.001), recurrent instability (12% vs 41%; P < 0.001), revision surgery (10% vs 24%; P = 0.015), osteoarthritis (OA) (5% vs 19%; P = 0.005), and restricted motion (161° vs 168°; P = 0.001). Forty‐one percent of patients had a “perfect outcome,” defined as ideal performance across all outcomes. Time from initial instability to presentation (odds ratio [OR] = 0.96; 95% confidence interval [CI], 0.92‐0.98; P = 0.006) and habitual/voluntary instability (OR = 0.17; 95% CI, 0.04‐0.77; P = 0.020) were negative predictors of achieving the “optimal observed outcome.” A predilection toward subluxations rather than dislocations before surgery (OR = 1.30; 95% CI, 1.02‐1.65; P = 0.030) was a positive predictor. Type of surgery performed was not a significant predictor. Conclusions After surgery for ASI, 64% of patients achieved the “optimal observed outcome” defined as minimal postoperative pain, no recurrent instability or OA, low revision surgery rates, and increased range of motion, of whom only 41% achieved a “perfect outcome.” Positive predictors were shorter time to presentation and predilection toward preoperative subluxations over dislocations. Level of Evidence Level IV, retrospective cohort.

AbstractPurposeIdentifying predictive factors for all‐cause reoperation after anterior cruciate ligament reconstruction could inform clinical decision making and improve risk mitigation. The primary purposes of this study are to (1) determine the incidence of all‐cause reoperation after anterior cruciate ligament reconstruction, (2) identify predictors of reoperation after anterior cruciate ligament reconstruction using machine learning methodology, and (3) compare the predictive capacity of the machine learning methods to that of traditional logistic regression.MethodsA longitudinal geographical database was utilized to identify patients with a diagnosis of new anterior cruciate ligament injury. Eight machine learning models were appraised on their ability to predict all‐cause reoperation after anterior cruciate ligament reconstruction. Model performance was evaluated via area under the receiver operating characteristics curve. To explore modeling interpretability and radiomic feature influence on the predictions, we utilized a game‐theory‐based method through SHapley Additive exPlanations.ResultsA total of 1400 patients underwent anterior cruciate ligament reconstruction with a mean postoperative follow‐up of 9 years. Two‐hundred and eighteen (16%) patients experienced a reoperation after anterior cruciate ligament reconstruction, of which 6% of these were revision ACL reconstruction. SHapley Additive exPlanations plots identified the following risk factors as predictive for all‐cause reoperation: diagnosis of systemic inflammatory disease, distal tear location, concomitant medial collateral ligament repair, higher visual analog scale pain score prior to surgery, hamstring autograft, tibial fixation via radial expansion device, younger age at initial injury, and concomitant meniscal repair. Pertinent negatives, when compared to previous studies, included sex and timing of surgery. XGBoost was the best‐performing model (area under the receiver operating characteristics curve of 0.77) and outperformed logistic regression in this regard.ConclusionsAll‐cause reoperation after anterior cruciate ligament reconstruction occurred at a rate of 16%. Machine learning models outperformed traditional statistics and identified diagnosis of systemic inflammatory disease, distal tear location, concomitant medial collateral ligament repair, higher visual analog scale pain score prior to surgery, hamstring autograft, tibial fixation via radial expansion device, younger age at initial injury, and concomitant meniscal repair as predictive risk factors for reoperation. Pertinent negatives, when compared to previous studies, included sex and timing of surgery. These models will allow surgeons to tabulate individualized risk for future reoperation for patients undergoing anterior cruciate ligament reconstruction.Level of evidenceIII.

Background: There is a significant long-term risk of posttraumatic osteoarthritis (PTOA) after anterior cruciate ligament reconstruction (ACLR). Elucidating the risk factors and successfully identifying at-risk patients is challenging. Purpose/Hypothesis: The purpose of this study was to produce machine learning survival models that can identify (1) patients at risk of symptomatic PTOA and (2) patients who are at risk of undergoing total knee arthroplasty (TKA) after ACLR. It was hypothesized that these models would outperform traditional Kaplan-Meier estimators. Study Design: Case-control study; Level of evidence, 3. Methods: A geographic database was used to identify patients undergoing ACLR between 1990 and 2016 with a minimum 7.5-year follow-up. Models were used to analyze various factors to predict the rate and time to (1) symptomatic osteoarthritis and (2) TKA using random survival forest (RSF) algorithms. Performance was measured using out-of-bag (OOB) c-statistic, calibration, and Brier score. The predictive performances of the RSF models were compared with Kaplan-Meier estimators. Model interpretability was enhanced utilizing global variable importance and partial dependence curves. Results: A total of 974 patients with ACLR and a minimum follow-up of 7.5 years were included; among these, 215 (22.1%) developed symptomatic osteoarthritis, and 25 (2.6%) progressed to TKA. The RSF algorithms achieved acceptable good to excellent predictive performance for symptomatic arthritis (OOB c-statistic, 0.75; Brier score, 0.128) and progression to TKA (OOB c-statistic, 0.89; Brier score, 0.026), respectively. Significant predictors of symptomatic PTOA included increased pain scores, older age, increased body mass index, increased time to ACLR, total number of arthroscopic surgeries before the diagnosis of arthritis, positive pivot-shift test after reconstruction, concomitant chondral injury, secondary meniscal tear, early (<250 days) or delayed (>500 days) return to sports or activity, and use of allograft. Significant predictors for TKA included older age, increased pain scores, total number of arthroscopic surgeries, high-demand activity/occupation, hypermobility, higher body mass index, systemic inflammatory disease, increased time to surgery, early (<250 days) or delayed (>500 days) return to sports or activity, and midsubstance tears. The Brier score over time revealed that RSF models outperformed traditional Kaplan-Meier estimators. Conclusion: Machine learning survival models were used to reliably identify patients at risk of developing symptomatic PTOA, and these models consistently outperformed traditional Kaplan-Meier estimators. Strong predictors for the development of PTOA after ACLR included increased pain scores at injury and postoperative visit, older age at injury, total number of arthroscopic procedures, positive postoperative pivot-shift test, and secondary meniscal tear.

Purpose To evaluate the cost‐effectiveness of 3 isolated meniscal repair (IMR) treatment strategies: platelet‐rich plasma (PRP)–augmented IMR, IMR with a marrow venting procedure (MVP), and IMR without biological augmentation. Methods A Markov model was developed to evaluate the baseline case: a young adult patient meeting the indications for IMR. Health utility values, failure rates, and transition probabilities were derived from the published literature. Costs were determined based on the typical patient undergoing IMR at an outpatient surgery center. Outcome measures included costs, quality‐adjusted life‐years (QALYs), and the incremental cost‐effectiveness ratio (ICER). Results Total costs of IMR with an MVP were $8,250; PRP‐augmented IMR, $12,031; and IMR without PRP or an MVP, $13,326. PRP‐augmented IMR resulted in an additional 2.16 QALYs, whereas IMR with an MVP produced slightly fewer QALYs, at 2.13. Non‐augmented repair produced a modeled gain of 2.02 QALYs. The ICER comparing PRP‐augmented IMR versus MVP‐augmented IMR was $161,742/QALY, which fell well above the $50,000 willingness‐to‐pay threshold. Conclusions IMR with biological augmentation (MVP or PRP) resulted in a higher number of QALYs and lower costs than non‐augmented IMR, suggesting that biological augmentation is cost‐effective. Total costs of IMR with an MVP were significantly lower than those of PRP‐augmented IMR, whereas the number of additional QALYs produced by PRP‐augmented IMR was only slightly higher than that produced by IMR with an MVP. As a result, neither treatment dominated over the other. However, because the ICER of PRP‐augmented IMR fell well above the $50,000 willingness‐to‐pay threshold, IMR with an MVP was determined to be the overall cost‐effective treatment strategy in the setting of young adult patients with isolated meniscal tears. Level of Evidence Level III, economic and decision analysis.

Background: Surgical and nonoperative management of anterior cruciate ligament (ACL) injuries seek to mitigate the risk of knee instability and secondary meniscal injury. However, the associated risk and timing of secondary meniscal tears have not been completely elucidated. Purpose: To compare risk and timing of secondary meniscal injury between patients receiving nonoperative management, delayed ACL reconstruction (ACLR), and early ACLR using a machine learning survival analysis. Study Design: Cohort study; Level of evidence, 3. Methods: A geographic database was used to identify and review records of patients with a diagnosis of ACL rupture between 1990 and 2016 with minimum 2-year follow-up. Patients undergoing ACLR were matched 1:1 with nonoperatively treated controls. Rate and time to secondary meniscal tear were compared using random survival forest algorithms; independent models were developed and internally validated for predicting injury-free duration in both cohorts. Performance was measured using out-of-bag c-statistic, calibration, and Brier score. Model interpretability was enhanced using global variable importance and partial dependence curves. Results: The study included 1369 patients who underwent ACLR and 294 patients who had nonoperative treatment. After matching, no significant differences in rates of secondary meniscal tear were found ( P = .09); subgroup analysis revealed the shortest periods of meniscal survival in patients undergoing delayed ACLR. The random survival forest algorithm achieved excellent predictive performance for the ACLR cohort, with an out-of-bag c-statistic of 0.80 and a Brier score of 0.11. Significant variables for risk of meniscal tear for the ACLR cohort included time to return to sports or activity ≤350 days, time to surgery ≥50 days, age at injury ≤40 years, and high-impact or rotational landing sports, whereas those in the nonoperative cohort model included time to RTS ≤200 days, visual analog scale pain score >3 at consultation, hypermobility, and noncontact sports. Conclusion: Delayed ACLR demonstrated the greatest long-term risk of meniscal injury compared with nonoperative treatment or early ACLR. Risk factors for decreased meniscal survival after ACLR included increased time to surgery, shorter time to return to sports or activity, older age at injury, and involvement in high-impact or rotational landing sports. Pending careful external validation, these models may be deployed in the clinical space to provide real-time insights and enhance decision making.

AbstractPurposeThis study sought to develop and internally validate a machine learning model to identify risk factors and quantify overall risk of secondary meniscus injury in a longitudinal cohort after primary ACL reconstruction (ACLR).MethodsPatients with new ACL injury between 1990 and 2016 with minimum 2‐year follow‐up were identified. Records were extensively reviewed to extract demographic, treatment, and diagnosis of new meniscus injury following ACLR. Four candidate machine learning algorithms were evaluated to predict secondary meniscus tears. Performance was assessed through discrimination using area under the receiver operating characteristics curve (AUROC), calibration, and decision curve analysis; interpretability was enhanced utilizing global variable importance plots and partial dependence curves.ResultsA total of 1187 patients underwent ACLR; 139 (11.7%) experienced a secondary meniscus tear at a mean time of 65 months post‐op. The best performing model for predicting secondary meniscus tear was the random forest (AUROC = 0.790, 95% CI: 0.785–0.795; calibration intercept = 0.006, 95% CI: 0.005–0.007, calibration slope = 0.961 95% CI: 0.956–0.965, Brier’s score = 0.10 95% CI: 0.09–0.12), and all four machine learning algorithms outperformed traditional logistic regression. The following risk factors were identified: shorter time to return to sport (RTS), lower VAS at injury, increased time from injury to surgery, older age at injury, and proximal ACL tear.ConclusionMachine learning models outperformed traditional prediction models and identified multiple risk factors for secondary meniscus tears after ACLR. Following careful external validation, these models can be deployed to provide real‐time quantifiable risk for counseling and timely intervention to help guide patient expectations and possibly improve clinical outcomes.Level of evidenceIII.

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.

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.