Gao Guanghan, Zhang Yaonan, Shi Lei, Wang Lin, Wang Fei, Xue Qingyun
Department of Orthopedics, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, China.
Department of Orthopedics, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, 100730, China.
Vis Comput Ind Biomed Art. 2025 Jun 6;8(1):14. doi: 10.1186/s42492-025-00195-w.
Knee osteoarthritis (KOA) is a prevalent chronic condition in the elderly and is often associated with instability caused by anterior cruciate ligament (ACL) degeneration. The functional integrity of ACL is crucial for the diagnosis and treatment of KOA. Radiographic imaging is a practical diagnostic tool for predicting the functional status of the ACL. However, the precision of the current evaluation methodologies remains suboptimal. Consequently, we aimed to identify additional radiographic features from X-ray images that could predict the ACL function in a larger cohort of patients with KOA. A retrospective analysis was conducted on 272 patients whose ACL function was verified intraoperatively between October 2021 and October 2024. The patients were categorized into ACL-functional and ACL-dysfunctional groups. Using least absolute shrinkage and selection operator regression and logistic regression, four significant radiographic predictors were identified: location of the deepest wear on the medial tibial plateau (middle and posterior), wear depth in the posterior third of the medial tibial plateau (> 1.40 mm), posterior tibial slope (PTS > 7.90°), and static anterior tibial translation (> 4.49 mm). A clinical prediction model was developed and visualized using a nomogram with calibration curves and receiver operating characteristic analysis to confirm the model performance. The prediction model demonstrated great discriminative ability, showing area under the curve values of 0.831 (88.4% sensitivity, 63.8% specificity) and 0.907 (86.1% sensitivity, 82.2% specificity) in the training and validation cohorts, respectively. Consequently, the authors established an efficient approach for accurate evaluation of ACL function in KOA patients.
膝关节骨关节炎（KOA）是老年人中常见的慢性疾病，常与前交叉韧带（ACL）退变导致的不稳定相关。ACL的功能完整性对于KOA的诊断和治疗至关重要。影像学检查是预测ACL功能状态的实用诊断工具。然而，当前评估方法的准确性仍不尽人意。因此，我们旨在从X线图像中识别出更多可预测更大队列KOA患者ACL功能的影像学特征。对2021年10月至2024年10月期间术中证实ACL功能的272例患者进行了回顾性分析。将患者分为ACL功能正常组和ACL功能障碍组。使用最小绝对收缩和选择算子回归及逻辑回归，确定了四个重要的影像学预测指标：内侧胫骨平台最深磨损部位（中部和后部）、内侧胫骨平台后三分之一处的磨损深度（>1.40mm）、胫骨后倾（PTS>7.90°）和静态胫骨前移（>4.49mm）。使用带有校准曲线的列线图和受试者工作特征分析开发并可视化了临床预测模型，以确认模型性能。该预测模型具有很强的鉴别能力，在训练队列和验证队列中的曲线下面积值分别为0.831（灵敏度88.4%，特异度63.8%）和0.907（灵敏度86.1%，特异度82.2%）。因此，作者建立了一种准确评估KOA患者ACL功能的有效方法。
