Hattori Masumi, Koyama Shuji
Department of Radiology, Tokai Memorial Hospital.
Nihon Hoshasen Gijutsu Gakkai Zasshi. 2004 Jul;60(7):993-9. doi: 10.6009/jjrt.kj00000922535.
Recent research has suggested that the measurement of regional atrophy in the structure of the medial temporal lobe is a promising way to discriminate Alzheimer-type dementia patients from healthy control subjects. The purpose of this study was to develop a technique to measure the medial temporal lobe automatically in axial CT images. Linear measurements of width of the inferior horns of the lateral ventricles, width of the medial temporal lobe, and the interuncal distance were performed. Area measurements of the inferior horns of the lateral ventricles were also performed. In the algorithm for the automatic measurement of the medial temporal lobe, brain contour, and sagittal plane were detected first, and the inferior horns of the lateral ventricles. Our method was applied to ten patients clearly without cerebral hemorrhage or infarct. The rates of accuracy of automated detection were 93% with the linear measurements and 75% with the area of the inferior horns. The rates were improved to 100% with the variable function of the threshold value. We suggest that this automated measurement method is both objective and simple enough to be used in routine clinical applications.
近期研究表明，测量内侧颞叶结构中的局部萎缩是区分阿尔茨海默型痴呆患者与健康对照者的一种有前景的方法。本研究的目的是开发一种在轴向CT图像中自动测量内侧颞叶的技术。对侧脑室下角宽度、内侧颞叶宽度和钩间距离进行了线性测量。还对侧脑室下角进行了面积测量。在自动测量内侧颞叶的算法中，首先检测脑轮廓和矢状面，以及侧脑室下角。我们的方法应用于十名明确无脑出血或梗死的患者。线性测量的自动检测准确率为93%，侧脑室下角面积的自动检测准确率为75%。通过阈值可变功能，准确率提高到了100%。我们认为这种自动测量方法既客观又简单，足以用于常规临床应用。
