Quantitative MRI is increasingly being used as a biomarker in neurological disorders. Cerebellar atrophy occurs in some Alternating Hemiplegia of Childhood (AHC) patients. However, it is not known if cerebellar atrophy can be a potential biomarker in AHC or if quantitative MRI is a reliable method to address this question. Here we determine the reproducibility of an MRI-volumetrics method to investigate brain volumes in AHC and apply it to a population of 14 consecutive AHC patients (ages 4-11 years). We studied method reproducibility in the first 11 patients and then performed correlation of cerebellar volumes, relative to published normal population means, with age in all 14. We used FreeSurfer 6.0.0 to automatically segment MRI images, then performed manual resegmentation correction by two different observers. No significant differences were observed in any of ten brain regions between the two reviewers: p > .591 and interclass Correlation Coefficient (ICC) ≥0.975 in all comparisons. Additionally, there were no significant differences between the means of the two reviewers and the automatic segmentation values: p ≥ .106 and ICC ≥0.994 in all comparisons. We found a negative correlation between cerebellar volume and age (R = -0.631, p = .037), even though only one patient showed any cerebellar atrophy upon formal readings of the MRIs by neuroradiology. Sample size did not allow us to rule out potential confounding variables. Thus, findings from this cross-sectional study should be considered as exploratory. Our study supports the prospective investigation of quantitative MRI-volumetrics of the cerebellum as a potential biomarker in AHC.

中文翻译：

---

儿童交替性偏瘫患者的磁共振成像体积分析：一项初步研究

定量 MRI 越来越多地被用作神经系统疾病的生物标志物。小脑萎缩发生在一些儿童交替性偏瘫 (AHC) 患者中。然而，尚不清楚小脑萎缩是否可以成为 AHC 的潜在生物标志物，或者定量 MRI 是否是解决这个问题的可靠方法。在这里，我们确定了 MRI 体积测量方法的可重复性，以研究 AHC 中的脑容量，并将其应用于 14 名连续的 AHC 患者（4-11 岁）。我们研究了前 11 名患者的方法可重复性，然后对所有 14 名患者的小脑体积（相对于已发表的正常人群平均值）与年龄进行了相关性分析。我们使用 FreeSurfer 6.0.0 自动分割 MRI 图像，然后通过两种不同的方法进行手动重新分割校正观察员。两位评论者之间的十个大脑区域中的任何一个都没有观察到显着差异：在所有比较中，p > .591 和类间相关系数 (ICC) ≥ 0.975。此外，两位评论者的平均值和自动分割值之间没有显着差异：在所有比较中，p ≥ .106 和 ICC ≥ 0.994。我们发现小脑体积与年龄之间呈负相关（R = -0.631，p = .037），即使只有一名患者在神经放射学的 MRI 正式读数时显示出任何小脑萎缩。样本量不允许我们排除潜在的混杂变量。因此，这项横断面研究的结果应被视为探索性的。我们的研究支持对小脑定量 MRI 体积测量作为 AHC 潜在生物标志物的前瞻性研究。