Citations

Abstract Purpose The purpose of this systematic review was to compare the inter‐ and intra‐observer reliability of Kellgren–Lawrence (KL) grading versus minimum joint space width (mJSW) measurement on knee radiographs, and to examine the influence of rater type, radiographic view and atlas‐based training. Methods A systematic search of PubMed/MEDLINE, Embase, Scopus and Web of Science was conducted through July 2025 following Preferred Reporting Items for Systematic reviews and Meta‐Analyses (PRISMA) guidelines. Studies reporting inter‐ and/or intra‐observer reliability for KL or mJSW on standard knee radiographs were included. Extracted data included kappa ( κ ) and intraclass correlation coefficient (ICC) values, rater expertise, radiographic view and atlas usage. Raters were categorized as expert, trainee/non‐expert or artificial intelligence (AI)‐assisted/automated. Results Twenty‐four studies (10,394 radiographs) met the inclusion criteria. Inter‐observer reliability was generally higher for mJSW (mean ICC = 0.82 ± 0.17) compared with KL (mean k = 0.72 ± 0.17), with similar trends observed for intra‐observer reliability. Experts and AI‐assisted systems outperformed trainees/non‐experts for both metrics. Reliability improved with semiflexed posteroanterior/Rosenberg views and atlas‐guided training. Conclusion This systematic review demonstrates that mJSW measurement generally shows higher reproducibility than KL grading. Reliability improved with expertise, standardized imaging and structured training, and AI‐assisted approaches performed comparably to expert raters. Quantitative mJSW should complement KL grading, particularly when combined with alignment measures, to enhance consistency and clinical relevance of radiographic assessment. Level of Evidence Level III.

Background: We aimed to develop automated data extraction pipelines with large language models (LLMs) to extract registry data from total hip arthroplasty (THA) operative notes and compare the performance with that of existing natural language processing (NLP) algorithms. Methods: We randomly sampled 1,000 primary THA cases from our institutional registry. Two human annotators manually reviewed each operative note for 3 data points: surgical approach, bearing surface, and fixation technique. All labeled THA notes were split into the development set (n = 239) and the testing set (n = 719). We developed a custom data extraction pipeline for each data point by combining an iteratively customized prompt with an LLM. The performance was compared with that of existing rules-based NLP algorithms. Results: The accuracy of LLMs was superior to that of NLP algorithms for all data points: surgical approach (96% compared with 94%), bearing surface (89% compared with 74%), and fixation technique (96% compared with 95%). Furthermore, the LLM accurately inferred the bearing surface for 80% of the notes that were ambiguous about the bearing surface. Conclusions: We developed LLM pipelines for extracting 3 registry-relevant data points from THA operative notes, demonstrating superior performance to existing NLP algorithms. Clinical Relevance: LLMs have the potential to impact clinical care, including the evaluation of electronic medical record free-text data. As registries serve as a cornerstone of orthopaedic evidence, this work demonstrates promise for LLMs to simplify, improve, and democratize the construction of registry databases from operative notes.

Artificial intelligence (AI) is increasingly shaping pediatric spine care, leveraging its rapid advancements in healthcare to improve efficiency, accuracy, and disease understanding. Moreover, machine learning and deep learning excel at detecting complex patterns. This holds promise in processing spinal deformity data, with the potential to surpass traditional statistical methods in predictive accuracy. Challenges persist, however, including unclear clinical implementation guidelines, limited model transparency, and ethical concerns surrounding data privacy and bias. Small sample sizes and the need for larger, diverse datasets further complicate integration. In order to realize AI’s transformative potential in pediatric spine care, these critical obstacles must be addressed for effective and ethical clinical adoption. This review examines the role of AI through applications such as image sorting, surgical outcome prediction, forecasting of spinal curve progression, and vertebral volumetric analysis using deep reasoning. It also explores possible intraoperative contributions from AI, including robotics and optimized screw trajectory planning, and the potential of large language models in clinical practice.

Background: Orthopaedic surgeons routinely delay surgical management of femoral neck fractures in patients taking direct oral anticoagulants (DOACs) to decrease perioperative bleeding and associated complications. However, this practice contradicts the principles of hip fracture management, as early surgery is associated with morbidity and mortality benefits. The purpose of this study was to quantify the association of DOAC use and perioperative outcomes in patients who underwent hemiarthroplasty (HA) or total hip arthroplasty (THA) for femoral neck fractures. We hypothesized that early surgical intervention on a patient taking a DOAC medication would not lead to worse perioperative outcomes. Methods: A retrospective cohort study was conducted on 2,833 patients who underwent primary THA or HA for femoral neck fractures between December 31, 2017, and January 29, 2024, across our hospital system. The patients taking a DOAC were divided into 3 groups based on the time since the last DOAC intake: 1 day, 2 days, and ≥3 days. Propensity matching was performed 1:1, accounting for age, sex, Elixhauser Comorbidity Index, preoperative chronic kidney disease stage, preoperative hemoglobin, body mass index, and hospital type. Subanalyses utilizing linear and conditional logistic regression models were performed to assess differences in outcomes between the groups that had a DOAC withheld and the control groups. Results: The mean age of all patients was 81 ± 10 years, 1,805 patients (64%) were women, and 207 patients (7%) were taking a DOAC prior to surgery. Despite comparable preoperative and postoperative hemoglobin levels between the groups that had a DOAC withheld and the control groups (all p > 0.05), the patients who had a DOAC withheld for 1 day were more likely to receive a postoperative blood transfusion (23.1% compared with 0%; p = 0.002). This difference in transfusion rate was not observed in other cohorts. There were no differences in medical complications, reoperation, discharge disposition, or mortality between the groups that had a DOAC withheld and the matched controls at any time point. Conclusions: Delaying surgical management due to DOAC medications may be unnecessary in patients undergoing arthroplasty for femoral neck fractures. Consideration should be given to adjusting transfusion triggers to reduce unwarranted blood transfusions in patients taking a DOAC. Level of Evidence: Therapeutic Level III . See Instructions for Authors for a complete description of levels of evidence.

Background/Objectives: Accurate assessment of spinal bone density is essential for evaluating bone health, particularly in preoperative planning. Conventional manual methods for Hounsfield unit (HU) measurements rely on single-slice measurements within the region of interest, limiting their precision and reproducibility in patients with severe vertebral deformities. We hypothesize that a novel deep-reasoning and learning model (DR-AI) can fully automate spinal bone density assessment volumetrically, with high correlations to spinal bone mineral density (BMD) obtained from dual-energy X-ray absorptiometry (DXA), as well as to the T- and Z-scores. Methods: A cross-sectional study was conducted on patients who had BMD assessment of their lumbar spine and lumbar CT scans within 1 year. The fully-automated DR model was utilized to analyze the soft-tissue window of the lumbar vertebrae CT scans. Spearman correlation coefficients were calculated to assess the strength of relationships between the computed volumetric HUs and the BMD, T-, and Z-scores from each vertebra. Results: 84 patients (67 females, mean age 74.1 ± 10.3 years; 17 males, mean age 68.1 ± 12.4 years) meeting inclusion criteria. Correlation analyses for L1 to L4 showed significant positive relationships (p < 0.0001), with the strongest correlation at L2 between HU and BMD (ρ = 0.75). Conclusions: the DR model for automated assessment of volumetric HUs offers a highly reliable, efficient, and precise alternative to DXA measurements.

Background: Nonoperative and operative management techniques after anterior cruciate ligament (ACL) injury are both appropriate treatment options for selected patients. However, the subsequent development of posttraumatic knee osteoarthritis (PTOA) remains an area of active study. Purpose: To compare the risk of PTOA between patients treated without surgery and with ACL reconstruction (ACLR) after primary ACL disruption using a machine learning causal inference model. Study Design: Cohort study; Level of evidence, 3. Methods: A geographic database identified patients undergoing ACLR between 1990 and 2016 with minimum 7.5-year follow-up. Variables collected include age, sex, body mass index, activity level, occupation, relevant comorbid diagnoses, radiographic findings, injury characteristics, and clinical course. Treatment effects of ACLR on the development of PTOA and progression to total knee arthroplasty (TKA) were analyzed with machine learning models (MLMs) in a causal inference estimator (targeted maximum likelihood estimation, TMLE), while controlling for confounders. Results: The study included 1194 patients with a minimum follow-up of 7.5 years, among whom 974 underwent primary reconstruction and 220 underwent nonoperative treatment. A total of 215 (22%) patients developed symptomatic PTOA in the ACLR group compared with 140 (64%) in the nonoperative treatment group ( P < .001), whereas 25 (3%) patients underwent TKA in the ACLR group compared with 50 (23%) in the nonoperative treatment group ( P < .001). Patients in the ACLR group had delayed TKA compared with patients in the nonoperative treatment group (193.4 vs 166.0 months, respectively; P = .02). TMLE evaluation revealed that reconstruction decreased the risk of PTOA by 11% (95% CI, 8%-13%; P < .001) compared with nonoperative treatment but did not demonstrate a significant effect on the rate of progression to TKA. Survival analysis with random forest algorithm demonstrated significant delay to the onset of PTOA as well as time to progression of TKA in patients undergoing ACLR. Additional risk factors for the development of PTOA, irrespective of treatment, included older age at injury, greater body mass index, total number of arthroscopic knee surgeries, and residual laxity at follow-up. Conclusion: MLMs in a causal inference estimator found ACLR to exert a significant treatment effect in reducing the rate of development of PTOA by 11% compared with nonoperative treatment. ACLR also delayed the onset of PTOA and progression to TKA.

Abstract Purpose Developing large‐scale, standardized radiographic registries for anterior cruciate ligament (ACL) injuries with artificial intelligence (AI) tools can enhance personalized orthopaedics. We propose deploying Artificial Intelligence for Knee Imaging Registration and Analysis (AKIRA), a trio of deep learning (DL) algorithms, to automatically classify and annotate radiographs. We hypothesize that algorithms can efficiently organize radiographs based on laterality, projection, identify implants and classify osteoarthritis (OA) grade. Methods A collection of 20,836 knee radiographs from all time points of treatment (mean orthopaedic follow‐up 70.7 months; interquartile range [IQR]: 6.8–172 months) were aggregated from 1628 ACL‐injured patients (median age 26 years [IQR: 19–42], 57% male). Three DL algorithms (EfficientNet, YOLO [You Only Look Once] and Residual Network) were employed. Radiograph laterality and projection (anterior‐posterior [AP], lateral, sunrise, posterior‐anterior, hip–knee–ankle and Camp‐Coventry intercondylar [notch]) were labelled by a DL model. Manually provided labels of metal fixation implants were used to develop a DL object detection algorithm. The degree of OA, both as measured by specific Kellgren–Lawrence (KL) grades, as well as based on a binarized label of OA (defined as KL Grade ≥2), on standing AP radiographs were classified using a DL algorithm. Individual model performances were evaluated on a subset of images prior to the deployment of AKIRA to registry construction using all ACL radiographs. Results The classification algorithms showed excellent performance in classifying radiographic laterality (F1 score: 0.962–0.975) and projection (F1 score: 0.941–1.0). The object detection algorithm achieved high precision–recall (area under the precision‐recall curve: 0.695–0.992) for identifying various metal fixations. The KL classifier reached concordances of 0.39–0.40, improving to 0.81–0.82 for binary OA labels. Sequential deployment of AKIRA following internal validation processed and labelled all 20,836 images with the appropriate views, implants, and the presence of OA within 88 min. Conclusion AKIRA effectively automated the classification and object detection in a large radiograph cohort of ACL injuries, creating an AI‐enabled radiographic registry with comprehensive details on laterality, projection, implants and OA. Study Design Cross‐sectional study. Level of Evidence Level IV.

AbstractDeep 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.

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.

The digitization of medical records and expanding electronic health records has created an era of “Big Data” with an abundance of available information ranging from clinical notes to imaging studies. In the field of rheumatology, medical imaging is used to guide both diagnosis and treatment of a wide variety of rheumatic conditions. Although there is an abundance of data to analyze, traditional methods of image analysis are human resource intensive. Fortunately, the growth of artificial intelligence (AI) may be a solution to handle large datasets. In particular, computer vision is a field within AI that analyzes images and extracts information. Computer vision has impressive capabilities and can be applied to rheumatologic conditions, necessitating a need to understand how computer vision works. In this article, we provide an overview of AI in rheumatology and conclude with a five step process to plan and conduct research in the field of computer vision. The five steps include (1) project definition, (2) data handling, (3) model development, (4) performance evaluation, and (5) deployment into clinical care.