Saniei, Sami

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.

As artificial intelligence (AI) advances in healthcare, encompassing robust applications for the diagnosis and prognostication of musculoskeletal diseases, clinicians must increasingly understand the implications of machine learning and deep learning in their practice. This review article explores computer vision algorithms and patient-specific, multimodal prediction models; provides a simple framework to guide discussion on the limitations of AI model development; and introduces the field of generative AI.

Abstract Morphological hip abnormalities (MHAs) significantly influence lifelong prognosis of the hip, contributing to early-onset osteoarthritis and impaired functionality. Developmental dysplasia of the hip (DDH) and femoroacetabular impingement (FAI) represent key pathologies, resulting from insufficient or excessive femoral head coverage, respectively. These abnormalities alter hip biomechanics, leading to structural damage, pain, and accelerated joint degeneration. Advances in genetic research have illuminated the interplay between genetics and mechanical loading in shaping hip morphology. Genes associated with osteoarthritis, DDH, and FAI include COL1A1, MMP13, and IL-6. Genes associated with FAI and osteoarthritis include ADAMTS4. Genes associated with DDH and osteoarthritis include FRZB, CX3CR1, ASPN, DKK1, PDRG1, GDF5, UQCC1, and TGF-β1. The mechanisms linking morphological derangements to symptomatic osteoarthritis remain incompletely understood. Multimodal approaches integrating imaging, biomechanics, and genetics may uncover distinct disease subtypes, enabling personalized interventions. Early detection of MHAs is critical in preventing early-onset osteoarthritis. Incorporating advanced imaging techniques, such as statistical shape modelling, can enhance the understanding of complex 3D hip morphologies and their progression to osteoarthritis. Future research should explore the genetic underpinnings of other morphologic hip conditions, including Slipped Capital Femoral Epiphysis and Legg–Calvé–Perthes disease, to refine preventive and therapeutic strategies. A comprehensive approach combining genetics, imaging, and clinical insights holds promise for mitigating the lifelong impact of MHAs.