Hentschel F, Besthorn C, Schmidt M H
Abteilung für Neuroradiologie, am Zentralinstitut für seelische Gesundheit, Mannheim.
Z Kinder Jugendpsychiatr Psychother. 1997 Dec;25(4):201-6.
Brain development is a self-organizing process and hence the brain structures correspond to a highly recursive system consisting of a multitude of parts. Quantitative analysis of recursive, self-similar structures is possible in CT scans as what is termed synergistic diagnostic using the fractal dimension. In 10 patients with anorexia nervosa (ICD-10, F50.0) we determined the fractal dimension of the cerebral cortex contour in a CT image using the "boxcounting method" after filtering of the data and separation with a defined cut-off. There was a significant difference between the results from CTs obtained before and after therapy. There was a high correlation between the fractal dimension and body weight, brain volume and sulcal cerebrospinal fluid volume, but only a low correlation with ventricular volume. The findings suggest that the qualification of the fractal dimension may be useful as one of postprocessing methods on CT scans. Further studies will be necessary to assess its importance for diagnosis.
大脑发育是一个自组织过程,因此大脑结构对应于一个由众多部分组成的高度递归系统。在CT扫描中,利用分形维数进行所谓的协同诊断,可以对递归、自相似结构进行定量分析。在10例神经性厌食症患者(国际疾病分类第10版,F50.0)中,我们在对数据进行滤波并以确定的截止值进行分离后,使用“盒计数法”在CT图像中确定大脑皮质轮廓的分形维数。治疗前后获得的CT结果之间存在显著差异。分形维数与体重、脑容量和脑沟脑脊液容量之间存在高度相关性,但与脑室容量的相关性较低。这些发现表明,分形维数的量化作为CT扫描的后处理方法之一可能是有用的。需要进一步研究以评估其对诊断的重要性。