Development and validation of a generalisable survival model to predict osteoarthritis progression

IF 2.8

Osteoarthritis and cartilage open Pub Date : 2025-09-25 DOI:10.1016/j.ocarto.2025.100688

H.H.T. Li , L.C. Chan , P.K. Chan , C. Wen

{"title":"Development and validation of a generalisable survival model to predict osteoarthritis progression","authors":"H.H.T. Li , L.C. Chan , P.K. Chan , C. Wen","doi":"10.1016/j.ocarto.2025.100688","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>To develop and validate a transfer-learning survival model to predict progression end-stage knee osteoarthritis (esKOA) or receive knee replacement (KR).</div></div><div><h3>Method</h3><div>A DeepSurv model was trained on the Osteoarthritis Initiative (OAI) dataset with baseline clinical variables, including age, sex, BMI, comorbidities, smoking status, prior knee injury and surgery, pain medication use, use of walking aids, and activity level (9560 knees). A generalisable model was then developed by fine-tuning the OAI-derived model with data from the Multicenter Osteoarthritis Study (MOST) Centre 1 (3002 knees). This model was validated on an independent dataset from MOST Centre 2 (2972 knees). Model performance was evaluated using the concordance index from 1000 bootstrap resamples. SHapley Additive exPlanations (SHAP) were employed to assess changes in feature importance after fine-tuning.</div></div><div><h3>Results</h3><div>The OAI-derived model performed well within OAI (C-index = 0.75) but fairly on MOST Centre 1 (C-index = 0.61, <em>p</em> < 0.0001), indicating domain shift or cross-cohort variation. Similarly, a model trained only on MOST Centre 1 data performed moderately within MOST (C-index = 0.63) but did not generalise to OAI (C-index = 0.60, <em>p</em> < 0.0001). After transfer learning, the generalised model maintained performance on OAI (C-index = 0.69) and improved on MOST Centre 1 (C-index = 0.64, <em>p</em> < 0.0001) and MOST Centre 2 (C-index = 0.67, <em>p</em> < 0.0001). SHAP analysis revealed that heart attack history, diabetes, smoking, and BMI became more influential predictors in the fine-tuned model.</div></div><div><h3>Conclusion</h3><div>Transfer learning enabled the development of a generalised model for knee OA prognosis that performs consistently across cohorts. By adapting to population-specific risk patterns, this approach enhances model generalisability and reduces bias from ethnic or demographic overrepresentation in training datasets.</div></div>","PeriodicalId":74377,"journal":{"name":"Osteoarthritis and cartilage open","volume":"7 4","pages":"Article 100688"},"PeriodicalIF":2.8000,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Osteoarthritis and cartilage open","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665913125001244","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Objective

To develop and validate a transfer-learning survival model to predict progression end-stage knee osteoarthritis (esKOA) or receive knee replacement (KR).

Method

A DeepSurv model was trained on the Osteoarthritis Initiative (OAI) dataset with baseline clinical variables, including age, sex, BMI, comorbidities, smoking status, prior knee injury and surgery, pain medication use, use of walking aids, and activity level (9560 knees). A generalisable model was then developed by fine-tuning the OAI-derived model with data from the Multicenter Osteoarthritis Study (MOST) Centre 1 (3002 knees). This model was validated on an independent dataset from MOST Centre 2 (2972 knees). Model performance was evaluated using the concordance index from 1000 bootstrap resamples. SHapley Additive exPlanations (SHAP) were employed to assess changes in feature importance after fine-tuning.

Results

The OAI-derived model performed well within OAI (C-index = 0.75) but fairly on MOST Centre 1 (C-index = 0.61, p < 0.0001), indicating domain shift or cross-cohort variation. Similarly, a model trained only on MOST Centre 1 data performed moderately within MOST (C-index = 0.63) but did not generalise to OAI (C-index = 0.60, p < 0.0001). After transfer learning, the generalised model maintained performance on OAI (C-index = 0.69) and improved on MOST Centre 1 (C-index = 0.64, p < 0.0001) and MOST Centre 2 (C-index = 0.67, p < 0.0001). SHAP analysis revealed that heart attack history, diabetes, smoking, and BMI became more influential predictors in the fine-tuned model.

Conclusion

Transfer learning enabled the development of a generalised model for knee OA prognosis that performs consistently across cohorts. By adapting to population-specific risk patterns, this approach enhances model generalisability and reduces bias from ethnic or demographic overrepresentation in training datasets.

查看原文本刊更多论文

预测骨关节炎进展的通用生存模型的开发和验证

目的建立并验证迁移学习生存模型，以预测终末期膝关节骨性关节炎（esKOA）的进展或接受膝关节置换术（KR）。方法DeepSurv模型在骨关节炎倡议（OAI）数据集上进行训练，该数据集具有基线临床变量，包括年龄、性别、BMI、合并症、吸烟状况、既往膝关节损伤和手术、止痛药使用、助行器使用和活动水平（9560个膝关节）。然后，利用多中心骨关节炎研究（MOST）中心1（3002个膝关节）的数据对oai衍生模型进行微调，建立了一个可推广的模型。该模型在来自MOST Centre 2（2972个膝关节）的独立数据集上进行了验证。使用来自1000个bootstrap样本的一致性指数来评估模型性能。采用SHapley加性解释（SHAP）来评估微调后特征重要性的变化。结果OAI衍生模型在OAI内表现良好（C-index = 0.75），但在MOST Centre 1上表现良好（C-index = 0.61, p < 0.0001），表明域转移或跨队列变异。同样，仅在MOST中心1数据上训练的模型在MOST范围内表现中等（C-index = 0.63），但不能推广到OAI （C-index = 0.60, p < 0.0001）。迁移学习后，广义模型在OAI上保持了良好的表现（C-index = 0.69），在MOST中心1 （C-index = 0.64, p < 0.0001）和MOST中心2 （C-index = 0.67, p < 0.0001）上有所改善。SHAP分析显示，心脏病发作史、糖尿病、吸烟和BMI在微调模型中成为更有影响力的预测因素。结论：迁移学习使膝关节OA预后的通用模型得以发展，该模型在整个队列中表现一致。通过适应特定人群的风险模式，该方法增强了模型的通用性，并减少了训练数据集中种族或人口统计学过度代表的偏见。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Osteoarthritis and cartilage open Orthopedics, Sports Medicine and Rehabilitation

CiteScore

3.30

自引率

0.00%

发文量