Computerized detection of vertebral compression fractures on lateral chest radiographs: preliminary results with a tool for early detection of osteoporosis.
Computerized detection of vertebral compression fractures on lateral chest radiographs: preliminary results with a tool for early detection of osteoporosis.
复制标题
DOI：
10.1118/1.2364053
复制
发表时间：
2006-12
期刊：
影响因子：
3.8
通讯作者：
S. Kasai;Feng Li;J. Shiraishi;Qiang Li;K. Doi
中科院分区：
文献类型：
--
作者：
S. Kasai;Feng Li;J. Shiraishi;Qiang Li;K. Doi
脊柱骨折(或椎体畸形)是骨质疏松症的常见后果，是世界上主要的公共卫生问题之一。脊椎骨折的早期发现很重要，因为及时的药物干预可以降低随后发生更多骨折的风险。胸部X光片通常用于检测肺病和心脏病，在侧位胸片上可以看到脊椎骨折。然而，研究人员注意到，在侧位胸片上可见的脊椎骨折中，约有50%被低估或漏报，即使骨折很严重。因此，我们的目标是开发一种计算机化的方法来检测侧位胸片上的椎体骨折，以辅助放射科医生的图像解释，从而实现对骨质疏松的早期诊断。这项研究中使用的病例是20名严重椎体骨折患者和118名非骨折患者，这一点得到了两位放射科医生的共识确认。放射科医生确定了骨折椎体的位置，并提供了椎体形状的形态测量数据，以评估计算机检测椎体终板的准确性。在我们的计算机化方法中，首先自动提取包括多个椎体终板的弯曲搜索区域，并将其拉直以使椎体终板成为水平方向。在矫正后的图像中使用水平线增强滤波器对边缘候选进行增强，然后使用多阈值技术和特征分析来识别脊椎终板。根据识别的椎体终板位置确定每个椎体的高度，并通过将测量的椎体高度与预期高度进行比较来检测骨折的椎体。我们的计算机方法检测骨折病例的灵敏度为95%(19/20)，每幅图像有1.03(139/135)个假阳性骨折。在形态计量学研究中，放射科医生识别椎体终板的准确率为76.6%(400/522)，用于训练的病例为70.9%(420/592)，用于测试的病例为70.9%(420/592)。我们准备了另外32个骨折病例进行验证测试，并检查了我们的计算机化方法的检测准确性。对这些病例的敏感性为75%(24/32)，每幅图像假阳性骨折数为1.03(33/32)。我们的初步结果表明，在侧位胸片上检测椎体骨折的自动计算机化方案具有辅助放射科医生检测椎体骨折的潜力。
Vertebral fracture (or vertebral deformity) is a very common outcome of osteoporosis, which is one of the major public health concerns in the world. Early detection of vertebral fractures is important because timely pharmacologic intervention can reduce the risk of subsequent additional fractures. Chest radiographs are used routinely for detection of lung and heart diseases, and vertebral fractures can be visible on lateral chest radiographs. However, investigators noted that about 50% of vertebral fractures visible on lateral chest radiographs were underdiagnosed or under-reported, even when the fractures were severe. Therefore, our goal was to develop a computerized method for detection of vertebral fractures on lateral chest radiographs in order to assist radiologists' image interpretation and thus allow the early diagnosis of osteoporosis. The cases used in this study were 20 patients with severe vertebral fractures and 118 patients without fractures, as confirmed by the consensus of two radiologists. Radiologists identified the locations of fractured vertebrae, and they provided morphometric data on the vertebral shape for evaluation of the accuracy of detecting vertebral end plates by computer. In our computerized method, a curved search area, which included a number of vertebral end plates, was first extracted automatically, and was straightened so that vertebral end plates became oriented horizontally. Edge candidates were enhanced by use of a horizontal line-enhancement filter in the straightened image, and a multiple thresholding technique, followed by feature analysis, was used for identification of the vertebral end plates. The height of each vertebra was determined from locations of identified vertebral end plates, and fractured vertebrae were detected by comparison of the measured vertebral height with the expected height. The sensitivity of our computerized method for detection of fracture cases was 95% (19/20), with 1.03 (139/135) false-positive fractures per image. The accuracy of identifying vertebral end plates, marked by radiologists in a morphometric study, was 76.6% (400/522) and 70.9% (420/592) for cases used for training and those for testing, respectively. We prepared 32 additional fracture cases for a validation test, and we examined the detection accuracy of our computerized method. The sensitivity for these cases was 75% (24/32) at 1.03 (33/32) false-positive fractures per image. Our preliminary results show that the automated computerized scheme for detecting vertebral fractures on lateral chest radiographs has the potential to assist radiologists in detecting vertebral fractures.