首页> 外文会议>Conference on computer-aided diagnosis >Knee osteoarthritis image registration: Data from the Osteoarthritis Initiative
【24h】

Knee osteoarthritis image registration: Data from the Osteoarthritis Initiative

机译:膝骨关节炎图像配准:骨关节炎计划的数据

获取原文

摘要

Knee osteoarthritis is a very common disease, in early stages, changes in joint structures are shown, some of the most common symptoms are; formation of osteophytes, cartilage degradation and joint space reduction, among others. Based on a joint space reduction measurement, Kellgren-Lawrence grading scale (K&L), is a very extensive used tool to asses radiological OA knee x-ray images, based on information obtained from these assessments, the objective of this work is to correlate the K&L score to the bilateral asymmetry between knees. Using public data from the Osteoarthritis initiative (OAI), a set of images with different K&L scores were used to determine a relationship of K&L score and the bilateral asymmetry, in order to measure the asymmetry between the knees, the right knee was registered to match the left knee, then a series of similarity metrics, mutual information, correlation, and mean squared error where computed to correlate the deformation (mismatch) of the knees to the K&L score. Radiological information was evaluated and scored by OAI radiologist groups. The results of the study suggest an association between Radiological K&L score and image registration metrics, mutual information and correlation is higher in the early stages, and mean squared error is higher in advanced stages. This association can be helpful to develop a computer aided grading tool.
机译:膝骨关节炎是一种非常常见的疾病,在早期阶段,显示出关节结构的改变,其中一些最常见的症状是:骨赘形成,软骨退化和关节间隙缩小等。基于联合空间缩减测量,Kellgren-Lawrence分级量表(K&L)是一种非常广泛的用于评估放射性OA膝关节X射线图像的工具,基于从这些评估中获得的信息,这项工作的目的是使K&L得分表示膝盖之间的双侧不对称性。根据骨关节炎倡议组织(OAI)的公开数据,使用一组具有不同K&L分数的图像来确定K&L分数与双侧不对称性之间的关系,以测量膝盖之间的不对称性,右膝被注册以匹配左膝,然后计算一系列相似度,互信息,相关性和均方误差,以将膝盖的变形(不匹配)与K&L得分相关联。放射学信息由OAI放射科医师小组评估和评分。研究结果表明,放射学K&L得分与图像配准指标之间存在关联,早期阶段的相互信息和相关性较高,晚期阶段的均方误差较高。这种关联可能有助于开发计算机辅助的评分工具。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号