Boniatis Ioannis, Cavouras Dionisis, Costaridou Lena, Kalatzis Ioannis, Panagiotopoulos Elias, Panayiotakis George
University of Patras, School of Medicine, Department of Medical Physics, 265 00 Patras, Greece.
Comput Biol Med. 2007 Dec;37(12):1786-95. doi: 10.1016/j.compbiomed.2007.05.005. Epub 2007 Jul 10.
A computer-based system was designed for the grading and quantification of hip osteoarthritis (OA) severity. Employing an active-contours segmentation model, 64 hip joint space (HJS) images (18 normal, 46 osteoarthritic) were obtained from the digitized radiographs of 32 unilateral and bilateral OA-patients. Shape features, generated from the HJS-images, and a hierarchical decision tree structure was used for the grading of OA. A shape features based regression model quantified the OA-severity. The system accomplished high accuracies in characterizing hips as "Normal" (100%), of "mild/moderate"-OA (93.8%) or "severe"-OA (96.7%). OA-severity values, as expressed by HJS-narrowing, correlated highly (r=0.9,p<0.001) with the values predicted by the regression model. The system may contribute to OA-patient management.
设计了一种基于计算机的系统，用于对髋骨关节炎（OA）严重程度进行分级和量化。利用主动轮廓分割模型，从32例单侧和双侧OA患者的数字化X线片上获取了64张髋关节间隙（HJS）图像（18张正常图像，46张骨关节炎图像）。从HJS图像生成形状特征，并使用分层决策树结构对OA进行分级。基于形状特征的回归模型对OA严重程度进行量化。该系统在将髋关节表征为“正常”（100%）、“轻度/中度”OA（93.8%）或“重度”OA（96.7%）方面具有很高的准确性。由HJS变窄表示的OA严重程度值与回归模型预测的值高度相关（r=0.9，p<0.001）。该系统可能有助于OA患者的管理。
