Farahibozorg S, Hashemi-Golpayegani S M, Ashburner J
Department of Biomedical Engineering, Amirkabir University of Technology, 424, Hafez Ave., 1591634311, Tehran, Iran,
Clin Neuroradiol. 2015 Mar;25(1):19-32. doi: 10.1007/s00062-013-0273-3. Epub 2014 Jan 12.
To observe age- and sex-related differences in the complexity of the global and hemispheric white matter (WM) throughout adulthood by means of fractal dimension (FD).
A box-counting algorithm was used to extract FD from the WM magnetic resonance images of 209 healthy adults from three structural layers, including general (gFD), skeleton (sFD), and boundaries (bFD). Model selection algorithms and statistical analyses, respectively, were used to examine the patterns and significance of the changes.
gFD and sFD showed inverse U-shape patterns with aging, with a slighter slope of increase from young to mid-age and a steeper decrease to the old. bFD was less affected by age. Sex differences were evident, specifically in gFD and sFD, with men showing higher FDs. Age × sex interaction was significant mainly in the hemispheric analysis, with men undergoing sharper age-related changes. After adjusting for the volume effect, age-related results remained approximately the same, but sex differences changed in most of the features, with women indicating higher values, specifically in the left hemisphere and boundaries. Right hemisphere was still more complex in men.
This study is the first that investigates the WM FD spanning adulthood, treating age both as a continuous and categorical variable. We found positive correlations between FD and volume, and our results show similarities with those investigating small-world properties of the brain networks, as well as those of functional complexity and WM integrity. These suggest that FD could yield a highly compact description of the structural changes and also might inform us about functional and cognitive variations.
通过分形维数（FD）观察成年期全脑和半球白质（WM）复杂性的年龄和性别差异。
采用盒计数算法从209名健康成年人的WM磁共振图像的三个结构层中提取FD，包括总体（gFD）、骨架（sFD）和边界（bFD）。分别使用模型选择算法和统计分析来检验变化的模式和显著性。
gFD和sFD随年龄呈倒U形模式，从青年到中年增加斜率较小，到老年下降更陡。bFD受年龄影响较小。性别差异明显，特别是在gFD和sFD方面，男性的FD值更高。年龄×性别交互作用主要在半球分析中显著，男性的年龄相关变化更明显。在调整体积效应后，年龄相关结果大致相同，但大多数特征的性别差异发生了变化，女性的值更高，特别是在左半球和边界。男性的右半球仍然更复杂。
本研究首次调查了跨越成年期的WM FD，将年龄视为连续和分类变量。我们发现FD与体积之间存在正相关，我们的结果与那些研究脑网络小世界属性、功能复杂性和WM完整性的研究结果相似。这些表明FD可以对结构变化进行高度紧凑的描述，也可能为我们提供有关功能和认知变化的信息。
