Machine Learning Algorithms: Prediction and Feature Selection for Clinical Refracture after Surgically Treated Fragility Fracture

Abstract

Background: The number of patients with fragility fracture has been increasing. Although the increasing number of patients with fragility fracture increased the rate of fracture (refracture), the causes of refracture are multifactorial, and its predictors are still not clarified. In this issue, we collected a registry-based longitudinal dataset that contained more than 7000 patients with fragility fractures treated surgically to detect potential predictors for clinical refracture. Methods: Based on the fact that machine learning algorithms are often used for the analysis of a large-scale dataset, we developed automatic prediction models and clarified the relevant features for patients with clinical refracture. Formats of input data containing perioperative clinical information were table data. Clinical refracture was documented as the primary outcome if the diagnosis of fracture was made at postoperative outpatient care. A decision-tree-based model, LightGBM, had moderate accuracy for the prediction in the test and the independent dataset, whereas the other models had poor accuracy or worse. Results: From a clinical perspective, rheumatoid arthritis (RA) and chronic kidney disease (CKD) were noted as the relevant features for patients with clinical refracture, both of which were associated with secondary osteoporosis. Conclusion: The decision-tree-based algorithm showed the precise prediction of clinical refracture, in which RA and CKD were detected as the potential predictors. Understanding these predictors may improve the management of patients with fragility fractures.

Osteoporosis and gut microbiota, radiofrequency echographic multispectrometry and machine learning: one year in review 2023

“…In fact, the presence of OP is often reported by the radiologist as a collateral finding on radiographs, thus being subject to interindividual variability. A similar problem occurs for the estimation of vertebral fractures with Genant semiquantitative method; in this regard, ML technologies are also useful in the creation of automated screening tools for vertebral fracture assessment and identification of risk factors for refractures (71)(72). Indeed, the…”

AI algorithms for accurate prediction of osteoporotic fractures in patients with diabetes: an up-to-date review

“…According to the presented data, most researches (18/19) of application of AI on prediction of fracture risks showed up after 2017, which is related to the vigorous and rapid development of AI algorithm in recent years. Researchers from various localities have gained universally acknowledged achievements in this domain, especially researchers from North America [ 36 , 39 , 44 , 48 , 49 , 51 ], Europe [ 41 – 43 , 45 – 47 , 50 , 52 , 53 ], and East Asia [ 37 , 38 , 40 ]. As for the application of algorithm patterns, supervised learning [ 41 – 51 , 53 , 54 ] is the most common pattern of risk prediction algorithm, and there are also reports on the application of unsupervised [ 36 , 52 ] learning methods to define high-risk groups of OFs.…”

“…In the field of fracture risk prediction, the application of AI is mainly divided into three types, among which the most common type is to improve the efficiency of fracture risk prediction by to establishing a new method or by applying the existing AI algorithm to the field of fracture risk prediction [ 38 – 40 , 42 – 44 , 46 – 49 , 53 , 54 ]. The following is to improve the prediction efficiency of the original ML prediction model or traditional prediction method by incorporating innovative imaging data or clinical characteristics [ 36 , 37 , 39 , 41 , 45 , 50 , 51 ]. Last but not least, to define the high-risk group [ 36 , 52 ] through ML algorithm, and then calculate its fracture prediction risk.…”