Citations

Abstract Artificial intelligence (AI) encompasses the development of systems that can perform human‐like tasks, such as treatment guidance, decision‐making, pattern recognition and understanding language. Within AI, machine learning and deep learning play pivotal roles in diagnosis and outcome prediction, while natural language processing aids in synthesising large datasets from the electronic medical record. In orthopaedics, AI has demonstrated success in various areas, including image evaluation, surgical planning, outcome prediction, cohort identification and administrative tasks. The purpose of this manuscript was to provide an overview of the benefits of AI implementation within the field of orthopaedics. An additional goal was to address the challenges associated with producing high quality AI‐based research in a rapidly developing field. Level of Evidence Level IV.

Objectives: Value based decision making regarding nonoperative management vs. early surgical stabilization for first-time anterior shoulder instability (ASI) events remains controversial. Thus, the purpose of this study was to perform 1) a systematic review of the current literature and 2) a Markov model-based cost-effectiveness analysis comparing an initial trial of nonoperative management to arthroscopic Bankart repair (ABR) for first-time ASI. Methods: A Markov Chain Monte Carlo probabilistic model (Figure 1) was developed to evaluate the outcomes and costs of 1,000 simulated patients (average age » 20 years old) with first-time ASI undergoing nonoperative management vs. ABR. Utility values, recurrence rates, and transition probabilities were derived from the published literature. Costs were determined based on the typical patient undergoing each treatment strategy at our institution. Outcome measures included costs, quality-adjust life-years (QALYs), and the incremental cost-effectiveness ratio (ICER). Probabilities of recurrence as well as the utility of each treatment strategy were obtained from the literature search as described above. Probabilities were pooled across studies to obtain a weighted average for both nonoperative treatment and ABR. Similarly, studies reporting scores on the Western Ontario Shoulder Instability Index (WOSI) scale were utilized to approximate shoulder-related quality of life after each treatment strategy and outcome. Costs for each treatment strategy and transition state in the model were considered from the payer perspective (commercial or government) and obtained from institutional data at a large, academic medical center. Results: The Markov model with Monte Carlo microsimulation demonstrated average (± standard deviation) costs for nonoperative management and ABR of $38,649 ± 10,521 and $43,052 ± 9,352, respectively. Total QALYS acquired over the 10-year time horizon were 7.7 ± 0.4 and 8.4 ± 0.5 for nonoperative management and ABR, respectively. The ICER comparing ABR to nonoperative management was found to be just $5,725/QALY, which falls substantially below the $50,000 willingness-to-pay (WTP) threshold (Figure 2). The mean number of recurrences were 2.6 ± 0.3 and 1.2 ± 0.2 for patients initially assigned to the nonoperative and ABR treatment groups, respectively. Out of 1,000 samples run over 1,000 trials, ABR was the optimal strategy in 98.7% of cases, with nonoperative management the optimal strategy in 1.3% of cases (Figure 3). Conclusions: The primary findings of this systematic review and cost-effectiveness analysis were as follows: 1) when compared to nonoperative management, arthroscopic Bankart repair (ABR) substantially reduces the risk of recurrent dislocations, 2) ABR results in greater shoulder function and higher WOSI scores than does nonoperative management, and 3) despite substantially higher upfront costs, ABR proved to be more cost-effective than an initial trial of nonoperative management, which can be attributed to both fewer recurrent dislocations and greater utility over the long-term.

Artificial intelligence (AI) has emerged as a transformative force in orthopedic surgery. Potentially encompassing pre-, intra-, and postoperative processes, it can process complex medical imaging, provide real-time surgical guidance, and analyze large datasets for outcome prediction and optimization. AI has shown improvements in surgical precision, efficiency, and patient outcomes across orthopedic subspecialties, and large language models and agentic AI systems are expanding AI utility beyond surgical applications into areas such as clinical documentation, patient education, and autonomous decision support. The successful implementation of AI in orthopedic surgery requires careful attention to validation, regulatory compliance, and healthcare system integration. As these technologies continue to advance, maintaining the balance between innovation and patient safety remains crucial, with the ultimate goal of achieving more personalized, efficient, and equitable healthcare delivery while preserving the essential role of human clinical judgment. This review examines the current landscape and future trajectory of AI applications in orthopedic surgery, highlighting both technological advances and their clinical impact. Studies have suggested that AI-assisted procedures achieve higher accuracy and better functional outcomes compared to conventional methods, while reducing operative times and complications. However, these technologies are designed to augment rather than replace clinical expertise, serving as sophisticated tools to enhance surgeons’ capabilities and improve patient care.

AbstractPurposeWhile several artificial intelligence (AI)‐assisted medical imaging applications are reported in the recent orthopaedic literature, comparison of the clinical efficacy and utility of these applications is currently lacking. The aim of this systematic review is to evaluate the effectiveness and reliability of AI applications in orthopaedic imaging, focusing on their impact on diagnostic accuracy, image segmentation and operational efficiency across various imaging modalities.MethodsBased on the PRISMA guidelines, a comprehensive literature search of PubMed, Cochrane and Scopus databases was performed, using combinations of keywords and MeSH descriptors ('AI', 'ML', 'deep learning', 'orthopaedic surgery' and 'imaging') from inception to March 2024. Included were studies published between September 2018 and February 2024, which evaluated machine learning (ML) model effectiveness in improving orthopaedic imaging. Studies with insufficient data regarding the output variable used to assess the reliability of the ML model, those applying deterministic algorithms, unrelated topics, protocol studies, and other systematic reviews were excluded from the final synthesis. The Joanna Briggs Institute (JBI) Critical Appraisal tool and the Risk Of Bias In Non‐randomised Studies‐of Interventions (ROBINS‐I) tool were applied for the assessment of bias among the included studies.ResultsThe 53 included studies reported the use of 11.990.643 images from several diagnostic instruments. A total of 39 studies reported details in terms of the Dice Similarity Coefficient (DSC), while both accuracy and sensitivity were documented across 15 studies. Precision was reported by 14, specificity by nine, and the F1 score by four of the included studies. Three studies applied the area under the curve (AUC) method to evaluate ML model performance. Among the studies included in the final synthesis, Convolutional Neural Networks (CNN) emerged as the most frequently applied category of ML models, present in 17 studies (32%).ConclusionThe systematic review highlights the diverse application of AI in orthopaedic imaging, demonstrating the capability of various machine learning models in accurately segmenting and analysing orthopaedic images. The results indicate that AI models achieve high performance metrics across different imaging modalities. However, the current body of literature lacks comprehensive statistical analysis and randomized controlled trials, underscoring the need for further research to validate these findings in clinical settings.Level of evidenceSystematic Review; Level of evidence IV.

Background: When injuring the anterior cruciate ligament (ACL), simultaneous meniscal tears often occur. However, little data describe ACL reconstruction (ACLR) and the subsequent risk of revision ACLR, as well as cartilaginous and meniscal damage after primary ACLR with different treatments for injury of the lateral meniscus (LM). Purpose: To investigate (1) how different LM treatments during primary ACLR affect the odds of revision ACLR and the cartilaginous and meniscal status at revision in a long-term follow-up and (2) whether the increased risk of subsequent meniscal and cartilaginous injuries seen with meniscal resection in the short term would continue in the long term, and (3) the increased risk with subsequent meniscal injuries with meniscal repair. Study Design: Cohort study; Level of evidence, 3. Methods: Data from the Swedish National Knee Ligament Registry were used. The study compared 2 groups: patients with isolated ACLR and patients with ACLR and concomitant LM injury, further subdividing the latter into 4 treatment types: repair, resection, both, and left in situ. Patients with medial meniscal injuries were excluded. The main outcomes studied were the odds for revision ACLR and the cartilaginous and meniscal status at revision 5 and 10 years after primary surgery. Results: Data for 5 years were available for 22,208 patients and data for 10 years were available for 11,058 patients. Compared with isolated ACLR, patients with concomitant LM injury left in situ had higher odds of revision ACLR at both 5 years (odds ratio [OR], 1.49; 95% CI, 1.14-1.95; P = .004) and 10 years (OR, 1.55; 95% CI, 1.09-2.19; P = .01), while those with LM repair had higher odds at 5 years (OR, 1.73; 95% CI, 1.23-2.43; P = .002). LM repair also increased the odds of subsequent meniscal injuries at 5 years (OR, 3.52; 95% CI, 1.76-7.03; P = .0004) and 10 years (OR, 7.26; 95% CI, 1.67-31.52; P = .008). Partial meniscectomy did not show an increased risk of cartilage injury, subsequent meniscal injury, or revision ACLR at long-term follow-up compared with the control group. Conclusion: This demonstrates that over 5 and 10 years, LM repair during primary ACLR is associated with an increased risk of revision ACLR and subsequent meniscal injuries. The higher risk of revision ACLR when the LM is left in situ emphasizes the need for careful consideration of treatment methods. Conversely, partial meniscectomy did not show an associated increased risk of cartilaginous injury in the long term and is associated with a lower risk of subsequent meniscal injury and revision ACLR than other treatment options, raising questions about whether preserving the meniscus at all costs is always the best approach.

AbstractArtificial intelligence (AI) has been influencing healthcare and medical research for several years and will likely become indispensable in the near future. AI is intended to support healthcare professionals to make the healthcare system more efficient and ultimately improve patient outcomes. Despite the numerous benefits of AI systems, significant concerns remain. Errors in AI systems can pose serious risks to human health, underscoring the critical need for safety, as well as adherence to ethical and moral standards, before these technologies can be integrated into clinical practice. To address these challenges, the development, certification, and deployment of medical AI systems must adhere to strict and transparent regulations. The European Commission has already established a regulatory framework for AI systems by enacting the European Union Artificial Intelligence Act. This review article, part of an AI learning series, discusses key considerations for medical AI systems such as reliability, accuracy, trustworthiness, lawfulness and legal compliance, ethical and moral alignment, sustainability, and regulatory oversight.Level of EvidenceLevel V.

Abstract Multimodal artificial intelligence (AI) has the potential to revolutionise healthcare by enabling the simultaneous processing and integration of various data types, including medical imaging, electronic health records, genomic information and real‐time data. This review explores the current applications and future potential of multimodal AI across healthcare, with a particular focus on orthopaedic surgery. In presurgical planning, multimodal AI has demonstrated significant improvements in diagnostic accuracy and risk prediction, with studies reporting an Area under the receiving operator curve presenting good to excellent performance across various orthopaedic conditions. Intraoperative applications leverage advanced imaging and tracking technologies to enhance surgical precision, while postoperative care has been advanced through continuous patient monitoring and early detection of complications. Despite these advances, significant challenges remain in data integration, standardisation, and privacy protection. Technical solutions such as federated learning (allowing decentralisation of models) and edge computing (allowing data analysis to happen on site or closer to site instead of multipurpose datacenters) are being developed to address these concerns while maintaining compliance with regulatory frameworks. As this field continues to evolve, the integration of multimodal AI promises to advance personalised medicine, improve patient outcomes, and transform healthcare delivery through more comprehensive and nuanced analysis of patient data. Level of Evidence: Level V.

Abstract Purpose The purpose of this article is to explore the potential of digital twin technologies in orthopaedics and to evaluate how their integration with artificial intelligence (AI) and deep learning (DL) can improve orthopaedic evaluation and treatment. This review addresses key applications of digital twins, including surgical planning, patient‐specific outcome prediction, augmented reality‐assisted surgery and simulation‐based surgical training. Methods Existing studies on digital twins in various domains, including engineering, biomedical and orthopaedics are reviewed. We also reviewed advancements in AI and DL relevant to digital twins. We focused on identifying key benefits, challenges and future directions for the implementation of digital twins in orthopaedic practice. Results The review highlights that digital twins offer significant potential to revolutionise orthopaedic care by enabling precise surgical planning, real‐time outcome prediction and enhanced training. Digital twins can model patient‐specific anatomy using advanced imaging techniques and dynamically update with real‐time data, providing valuable insights during surgery and postoperative care. However, challenges such as the need for large‐scale data sets, technological limitations and integration issues must be addressed to fully realise these benefits. Conclusion Digital twins represent a promising frontier in orthopaedic research and practice, with the potential to improve patient outcomes and enhance surgical precision. To enable widespread adoption, future research must focus on overcoming current challenges and further refining the integration of digital twins with AI and DL technologies. Level of Evidence Level V.

AbstractPurposeTo systematically review the literature regarding machine learning in leg length discrepancy (LLD) and to provide insight into the most relevant manuscripts on this topic in order to highlight the importance and future clinical implications of machine learning in the diagnosis and treatment of LLD.MethodsA systematic electronic search was conducted using PubMed, OVID/Medline and Cochrane libraries in accordance with Preferred Reporting Items for Systematic Review and Meta‐Analysis guidelines. Two observers independently screened the abstracts and titles of potential articles.ResultsA total of six studies were identified in the search. All measurements were calculated using standardized anterior‐posterior long‐leg radiographs. Five (83.3%) of the studies used measurements of the femoral length, tibial length and leg length to assess LLD, whereas one (16.6%) study used the iliac crest height difference to quantify LLD. The deep learning models showed excellent reliability in predicting all length measurements with intraclass correlation coefficients ranging from 0.98 to 1.0 and mean absolute error (MAE) values ranging from 0.11 to 0.45 cm. Three studies reported measurements of LLD, and the convolutional neural network model showed the lowest MAE of 0.13 cm in predicting LLD.ConclusionsMachine learning models are effective and efficient in determining LLD. Implementation of these models may reduce cost, improve efficiency and lead to better overall patient outcomes.Clinical RelevanceThis review highlights the potential of deep learning (DL) algorithms for accurate and reliable measurement of lower limb length and leg length discrepancy (LLD) on long‐leg radiographs. The reported mean absolute error and intraclass correlation coefficient values indicate that the performance of the DL models was comparable to that of radiologists, suggesting that DL‐based assessments could potentially be used to automate the measurement of lower limb length and LLD in clinical practice.Level of EvidenceLevel IV.