基于谐波分析法观察抑郁症患者脑网络形态学改变

目的 评估基于谐波分析法观察抑郁症(DD)患者脑网络形态学改变的可行性。方法 对55例DD患者(DD组)和46名正常对照者(NC组)采集全脑3D高分辨T1WI,利用FreeSurfer 5.3.0工具构建脑网络6种形态学特征，包括脑区顶点数、表面积、灰质体积、平均皮层厚度、高斯曲率及折叠指数，以拉普拉斯矩阵特征分解获得公共谐波；比较组间不同形态学特征所在谐波能量及脑区灰质体积。结果 组间不同形态学特征总谐波能量差异均无统计学意义(P均>0.05)。组间特定谐波能量差异有统计学意义(P均<0.05),主要包括脑区顶点数谐波能量差异为第2、6、15、44、57谐波，表面积谐波能量差异为第2、6、16、57谐波，灰质体积谐波能量差异为第2、12、13、15、57谐波，平均皮层厚度谐波能量差异为第2、19、35、36、44谐波，高斯曲率谐波能量差异为第34、40、54、57谐波，以及折叠指数谐波能量差异为第5、16、21、57谐波。DD组左脑区颞横皮质区域灰质体积显著大于NC组(t=2.900,P=0.004)。结论 谐波分析法可用于观察DD患者脑网络形态学改变。

Harmonic waves analysis for observing morphological brain network changes in depressive disorder patients

Objective To explore the feasibility of harmonic waves analysis for observing morphological brain network changes in patients with depressive disorder (DD). Methods Whole brain 3D high resolution T1WI of 55 DD patients (DD group) and 46 normal controls (NC group) were acquired. Six kinds of morphological features brain network were constructed with FreeSurfer tool, including the number of brain region vertices, surface area, gray matter volume, average cortical thickness, Gaussian curvature and fold index. Laplace operator was applied to obtain common harmonic wave. The harmonic power of different morphological features and the gray matter volume in different brain regions were compared between groups. Results No significant difference of total harmonic energy was found between groups. The specific harmonic wave energies were significantly different between groups, including the number of brain region vertices corresponding to the 2nd, 6th, 15th, 44th and 57th harmonic waves, surface area corresponding to the 2nd, 6th, 16th and 57th harmonic waves, gray matter volume corresponding to the 2nd, 12th, 13th, 15th and 57th harmonic waves, average cortical thickness corresponding to the 2nd, 19th, 35th, 36th and 44th harmonic waves, Gaussian curvature corresponding to the 34th, 40th, 54th and 57th harmonic waves, as well as fold index corresponding to the 5th, 16th, 21st and 57th harmonic waves. Gray matter volumes of transverse temporal gyrus in left hemisphere in DD group were significantly larger than that in NC group (t=2.900, P=0.004). Conclusion Harmonic waves analysis was feasible for observing morphological brain network changes in DD patients.