Artificial Intelligence in Orthopedics. Where are We and Where are We Going?

Artificial intelligence (AI) has emerged as one of the most promising tools in the transformation of the healthcare sector, offering innovative solutions that support diagnosis, as well as guiding the treatment and rehabilitation of patients.[1] In general terms, AI refers to computational systems capable of performing tasks that usually require human intelligence, such as pattern recognition, learning from data, and decision-making.[2] Within the healthcare field, its impact has been significant in areas such as radiology, robotic surgery, and the optimization of workflows in hospitals.

Current Advances in Artificial Intelligence in Traumatology

In the field of traumatology, AI has begun to play a key supporting role in areas such as diagnostic imaging, predictive analysis of postoperative complications, robotic surgery, and medical training.[1] [2] [3] [4] [5] [Figure 1]

Fig. 1 Artificial Intelligence in Traumatology.

AI in Fracture Detection

Regarding fracture detection, AI algorithms have demonstrated superior diagnostic capability in numerous studies, including the detection of vertebral,[6] [7] humeral,[8] femoral,[9] [10] shoulder,[11] [12] elbow,[13] [14] and ankle fractures,[15] mostly based on X-rays. Both Shen[6] and Zhang,[7] using trained algorithms, demonstrated sensitivity of 83–95%, specificity of 94–98%, and accuracy of 96–97% for the detection of vertebral fractures. Chung[8], by training a convolutional neural network (CNN) on 1,891 X-rays, reported a sensitivity of 99% and a specificity of 97% in the detection of humeral fractures. Beyaz,[9] by training a CNN, demonstrated a sensitivity of 82%, specificity of 72%, and an accuracy of 79% in the detection of femoral fractures. In shoulder cases, Uysal[11] reported a diagnostic accuracy of 84% for fracture detection through the training of two ensemble models. Rayan[13] used a CNN model trained on 58,817 X-rays for detecting elbow fractures in the pediatric population, achieving an accuracy of 88%, sensitivity of 91%, and specificity of 84%. Finally, Ashkani-Esfahani[15] demonstrated a sensitivity of 99%, specificity of 99%, and accuracy of 99% with a Transfer Learning algorithm based on InceptionV3 for the detection of ankle fractures.

AI in Postoperative Predictive Analysis

Artificial intelligence can support medical decision-making by interpreting complex analyses as predictors of postoperative complication risk, guiding more personalized clinical management. This analysis takes into account factors associated with each patient (age, sex, associated pathologies, among others), genetic information, and imaging,[16] and can predict patient outcomes. Numerous studies have been published using Machine Learning (ML), which predict the rate of postoperative complications in adults undergoing spine surgery[17] and arthroscopic hip preservation surgery.[18] In knee surgery, studies have been able to predict functional outcomes in patients undergoing osteochondral transplantation,[19] osteoarthritis progression toward prosthetic surgery,[20] and the need for hospitalization after anterior cruciate ligament reconstruction surgery.[21]

AI in Robotic Surgery

Robotic surgery leverages the advantages of complex computational calculations to optimize surgical outcomes, achieving greater precision in implant positioning and the intraoperative reduction of fractures in trauma settings. Additionally, the image-based computational planning that underpins robotic surgery allows for more precise bone cuts in the surgical area, optimizing the restoration of limb biomechanics and achieving better soft tissue tension to prevent failure.[22]

The greatest number of cases and demonstrated benefits of robotic surgery have been in lower limb arthroplasty, representing 90% of the market.[22] It has been shown that the use of this technology improves the accuracy of implant positioning in both hip and knee surgeries, although the evidence is less clear regarding functional outcomes or long-term implant survival.[23] [24]

There have also been advances in spinal surgery, ranging from surgical planning to the use of robotic arms with augmented reality.[25] [26]

AI in Medical Training

The application of AI in surgical clinical simulators uses immersive reality with automated anatomical visualization, providing users with enhanced surgical experiences and feedback to correct technical errors.[27] The benefits of this training modality have been demonstrated, achieving improved technical precision for both residents and surgeons.[28] [29] Other advantages of using this technology include lower financial costs and reduced radiation exposure compared to training with cadavers.[30]

Challenges and Future of AI in Traumatology

While progress is notable, the implementation of AI in clinical practice faces significant challenges. A recent article published by the ISAKOS Young Professionals Task Force[31] showed that only 25% of respondents use AI in clinical practice.

Furthermore, the interpretability of learning algorithms remains an obstacle, as physicians need to understand how predictions are generated to trust them. Furthermore, the integration of these technologies into hospital systems requires adequate infrastructure and staff training. Ethical questions also arise regarding liability in the event of diagnostic or surgical errors resulting from the use of AI.

Conclusion

For AI to become a standard support tool in traumatology, it is essential to adopt a proactive approach to its integration. Training in digital technologies and the development of collaborations between physicians, engineers, and data scientists will be key to maximizing the benefits of AI. Additionally, it is crucial to establish ethical regulations and rigorous validation protocols to ensure the safety and effectiveness of these tools in the clinical setting.

No conflict of interest has been declared by the author(s).

• Referencias

• 1 Khoriati AA, Shahid Z, Fok M. et al. Artificial intelligence and the orthopaedic surgeon: A review of the literature and potential applications for future practice: Current concepts. J ISAKOS 2024; 9 (02) 227-233
• 2 Farhadi F, Barnes MR, Sugito HR, Sin JM, Henderson ER, Levy JJ. Applications of artificial intelligence in orthopaedic surgery. Front Med Technol 2022; 4: 995526
• 3 Lex JR, Di Michele J, Koucheki R, Pincus D, Whyne C, Ravi B. Artificial Intelligence for Hip Fracture Detection and Outcome Prediction: A Systematic Review and Meta-analysis. JAMA Netw Open 2023; 6 (03) e233391
• 4 Tieu A, Kroen E, Kadish Y. et al. The Role of Artificial Intelligence in the Identification and Evaluation of Bone Fractures. Bioengineering (Basel) 2024; 11 (04) 338
• 5 Lo Mastro A, Grassi E, Berritto D. et al. Artificial intelligence in fracture detection on radiographs: a literature review. Jpn J Radiol 2025; 43 (04) 551-585
• 6 Shen L, Gao C, Hu S. et al. Using Artificial Intelligence to Diagnose Osteoporotic Vertebral Fractures on Plain Radiographs. J Bone Miner Res 2023; 38 (09) 1278-1287
• 7 Zhang J, Liu F, Xu J. et al. Automated detection and classification of acute vertebral body fractures using a convolutional neural network on computed tomography. Front Endocrinol (Lausanne) 2023; 14: 1132725
• 8 Chung SW, Han SS, Lee JW. et al. Automated detection and classification of the proximal humerus fracture by using deep learning algorithm. Acta Orthop 2018; 89 (04) 468-473
• 9 Beyaz S, Açıcı K, Sümer E. Femoral neck fracture detection in X-ray images using deep learning and genetic algorithm approaches. Jt Dis Relat Surg 2020; 31 (02) 175-183
• 10 Oakden-Rayner L, Gale W, Bonham TA. et al. Validation and algorithmic audit of a deep learning system for the detection of proximal femoral fractures in patients in the emergency department: a diagnostic accuracy study. Lancet Digit Health 2022; 4 (05) e351-e358
• 11 Uysal F, Hardalaç F, Peker O, Tolunay T, Tokgöz N. Classification of Shoulder X-Ray Images with Deep Learning Ensemble Models. Appl Sci (Basel) 2021; 11: 2723
• 12 Magnéli M, Ling P, Gislén J. et al. Deep learning classification of shoulder fractures on plain radiographs of the humerus, scapula and clavicle. PLoS One 2023; 18 (08) e0289808
• 13 Rayan JC, Reddy N, Kan JH, Zhang W, Annapragada A. Binomial Classification of Pediatric Elbow Fractures Using a Deep Learning Multiview Approach Emulating Radiologist Decision Making. Radiol Artif Intell 2019; 1 (01) e180015
• 14 Luo J, Kitamura G, Arefan D, Doganay E, Panigrahy A, Wu S. Knowledge-Guided Multiview Deep Curriculum Learning for Elbow Fracture Classification. Mach Learn Med Imaging 2021; 12966: 555-564
• 15 Ashkani-Esfahani S, Mojahed Yazdi R, Bhimani R. et al. Detection of ankle fractures using deep learning algorithms. Foot Ankle Surg 2022; 28 (08) 1259-1265
• 16 Hashimoto DA, Rosman G, Rus D, Meireles OR. Artificial Intelligence in Surgery: Promises and Perils. Ann Surg 2018; 268 (01) 70-76
• 17 Kim JS, Arvind V, Oermann EK. et al. Predicting Surgical Complications in Patients Undergoing Elective Adult Spinal Deformity Procedures Using Machine Learning. Spine Deform 2018; 6 (06) 762-770
• 18 Kunze KN, Polce EM, Clapp I, Nwachukwu BU, Chahla J, Nho SJ. Machine Learning Algorithms Predict Functional Improvement After Hip Arthroscopy for Femoroacetabular Impingement Syndrome in Athletes. J Bone Joint Surg Am 2021; 103 (12) 1055-1062
• 19 Ramkumar PN, Karnuta JM, Haeberle HS. et al. Association Between Preoperative Mental Health and Clinically Meaningful Outcomes After Osteochondral Allograft for Cartilage Defects of the Knee: A Machine Learning Analysis. Am J Sports Med 2021; 49 (04) 948-957
• 20 Pareek A, Parkes CW, Bernard CD, Abdel MP, Saris DBF, Krych AJ. The SIFK score: a validated predictive model for arthroplasty progression after subchondral insufficiency fractures of the knee. Knee Surg Sports Traumatol Arthrosc 2020; 28 (10) 3149-3155
• 21 Lu Y, Forlenza E, Cohn MR. et al. Machine learning can reliably identify patients at risk of overnight hospital admission following anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2021; 29 (09) 2958-2966
• 22 Innocenti B, Bori E. Robotics in orthopaedic surgery: why, what and how?. Arch Orthop Trauma Surg 2021; 141 (12) 2035-2042
• 23 Ruangsomboon P, Ruangsomboon O, Pornrattanamaneewong C, Narkbunnam R, Chareancholvanich K. Clinical and radiological outcomes of robotic-assisted versus conventional total knee arthroplasty: a systematic review and meta-analysis of randomized controlled trials. Acta Orthop 2023; 94: 60-79
• 24 Chen X, Xiong J, Wang P. et al. Robotic-assisted compared with conventional total hip arthroplasty: systematic review and meta-analysis. Postgrad Med J 2018; 94 (1112) 335-341
• 25 Volk VL, Steele KA, Cinello-Smith M. et al. Pedicle Screw Placement Accuracy in Robot-Assisted Spinal Fusion in a Multicenter Study. Ann Biomed Eng 2023; 51 (11) 2518-2527
• 26 Groisser BN, Thakur A, Hillstrom HJ. et al. Fully automated determination of robotic pedicle screw accuracy and precision utilizing computer vision algorithms. J Robot Surg 2024; 18 (01) 278
• 27 Park JJ, Tiefenbach J, Demetriades AK. The role of artificial intelligence in surgical simulation. Front Med Technol 2022; 4: 1076755
• 28 Kuhn AW, Yu JK, Gerull KM, Silverman RM, Aleem AW. Virtual Reality and Surgical Simulation Training for Orthopaedic Surgery Residents: A Qualitative Assessment of Trainee Perspectives. JBJS Open Access 2024; 9 (01) e23
• 29 Schöbel T, Schuschke L, Youssef Y, Rotzoll D, Theopold J, Osterhoff G. Immersive virtual reality in orthopedic surgery as elective subject for medical students : First experiences in curricular teaching. Orthopadie (Heidelb) 2024; 53 (05) 369-378
• 30 Gomindes AR, Adeeko ES, Khatri C. et al. Use of Virtual Reality in the Education of Orthopaedic Procedures: A Randomised Control Study in Early Validation of a Novel Virtual Reality Simulator. Cureus 2023; 15 (09) e45943
• 31 Ghassemi M, Naumann T, Schulam P, Beam AL, Chen IY, Ranganath R. A Review of Challenges and Opportunities in Machine Learning for Health. AMIA Jt Summits Transl Sci Proc 2020; 2020: 191-200

Publication History

Article published online:
20 May 2025