Dataset related to article "Preoperative diagnosis of periprosthetic infection in patients undergoing hip or knee revision arthroplasties: development and validation of machine learning algorithm".

This record contains raw data related to article "Preoperative diagnosis of periprosthetic infection in patients undergoing hip or knee revision arthroplasties: development and validation of machine learning algorithm" Abstract: Background: Periprosthetic joint infection (PJI) following total hip and knee arthroplasty remains an extremely challenging and relatively high complication. This study aims to develop, validate and evaluate the use of machine learning (ML) algorithm to predict PJI in patients undergoing revision arthroplasties. Methods: A comprehensive review of patients undergoing hip or knee revision arthroplasty from 1 January 2015 to 31 March 2021 was conducted. Clinical data coming from preoperative patients history, laboratory analysis and demographic characteristics of patients were screened. Final data have been used to train a Logistic Regression model with the aim of predicting PJI preoperatively. Results: 1360 patients were enrolled, 1141 in the aseptic cohort and 219 in the infected cohort were included. ML demonstrated good discriminatory performance in predicting PJI in the selected patients (area under the curve 0.770 ± 0.006 in the training set and 0.720 ± 0.057 in the test set), and identified 3 significant predictors of PJI. Conclusion: ML algorithm trained using preoperative clinical data accurately predicted PJI. The incorporation of ML models into preoperative assessment of patients undergoing prosthetic revision procedures are useful in providing specific risk assessment to aid individualised counselling, shared decision making and presurgical optimization. Keywords: Artificial intelligence; Arthroplasty; Deep machine learning; Hip; Knee; Periprosthetic joint infection.本数据集包含与论文《髋或膝关节翻修术患者假体周围感染的术前诊断：机器学习算法的开发与验证》相关的原始数据。 摘要：背景：全髋关节与膝关节置换术后假体周围关节感染（Periprosthetic Joint Infection, PJI）仍是极具挑战性且并发症发生率相对较高的临床难题。本研究旨在开发、验证并评估机器学习（Machine Learning, ML）算法在接受翻修关节置换术患者中预测PJI的应用价值。 方法：本研究回顾性收集2015年1月1日至2021年3月31日期间接受髋或膝关节翻修关节置换术患者的临床资料，筛选患者术前病史、实验室检验结果及人口统计学特征等相关数据，最终使用筛选后的数据集训练逻辑回归模型，以实现术前假体周围关节感染的预测。 结果：本研究共纳入1360例患者，其中无菌性队列1141例，感染性队列219例。机器学习模型在目标患者的PJI预测中展现出良好的区分性能：训练集曲线下面积（Area Under the Curve, AUC）为0.770 ± 0.006，测试集为0.720 ± 0.057，并筛选出3项具有统计学意义的PJI预测因子。 结论：基于术前临床数据训练的机器学习算法可精准预测假体周围关节感染。将机器学习模型纳入假体翻修术前评估流程，可为患者提供个性化的风险评估，有助于开展个体化咨询、共同决策及术前优化管理。 关键词：人工智能（Artificial Intelligence）；关节置换术（Arthroplasty）；深度机器学习（Deep Machine Learning）；髋（Hip）；膝（Knee）；假体周围关节感染（Periprosthetic Joint Infection）。