Department of Child Psychiatry and Rehabilitation, Institute of Maternity and Child Medical Research, Affiliated Shenzhen Maternity & Child Healthcare Hospital, Southern Medical University, Shenzhen, China.
Department of Radiology, Shenzhen Kangning Hospital, Shenzhen, China.
Psychol Med. 2024 Mar;54(4):710-720. doi: 10.1017/S0033291723002386. Epub 2023 Aug 29.
Obsessive-compulsive disorder (OCD) is a classic disorder on the compulsivity spectrum, with diverse comorbidities. In the current study, we sought to understand OCD from a dimensional perspective by identifying multimodal neuroimaging patterns correlated with multiple phenotypic characteristics within the striatum-based circuits known to be affected by OCD.
Neuroimaging measurements of local functional and structural features and clinical information were collected from 110 subjects, including 51 patients with OCD and 59 healthy control subjects. Linked independent component analysis (LICA) and correlation analysis were applied to identify associations between local neuroimaging patterns across modalities (including gray matter volume, white matter integrity, and spontaneous functional activity) and clinical factors.
LICA identified eight multimodal neuroimaging patterns related to phenotypic variations, including three related to symptoms and diagnosis. One imaging pattern (IC9) that included both the amplitude of low-frequency fluctuation measure of spontaneous functional activity and white matter integrity measures correlated negatively with OCD diagnosis and diagnostic scales. Two imaging patterns (IC10 and IC27) correlated with compulsion symptoms: IC10 included primarily anatomical measures and IC27 included primarily functional measures. In addition, we identified imaging patterns associated with age, gender, and emotional expression across subjects.
We established that data fusion techniques can identify local multimodal neuroimaging patterns associated with OCD phenotypes. The results inform our understanding of the neurobiological underpinnings of compulsive behaviors and OCD diagnosis.
强迫症（OCD）是强迫谱系中的一种经典障碍，伴有多种共病。在目前的研究中，我们试图从多维的角度理解 OCD，通过识别与已知受 OCD 影响的纹状体为基础的回路中的多个表型特征相关的多模态神经影像学模式。
从 110 名受试者中收集局部功能和结构特征的神经影像学测量值和临床信息，包括 51 名 OCD 患者和 59 名健康对照者。应用链接独立成分分析（LICA）和相关分析来识别跨模态（包括灰质体积、白质完整性和自发功能活动）的局部神经影像学模式与临床因素之间的关联。
LICA 确定了 8 个与表型变化相关的多模态神经影像学模式，包括与症状和诊断相关的 3 个模式。一个包含自发功能活动的低频波动幅度的测量值和白质完整性测量值的成像模式（IC9）与 OCD 诊断和诊断量表呈负相关。两个与强迫症状相关的成像模式（IC10 和 IC27）：IC10 主要包括解剖学测量值，而 IC27 主要包括功能测量值。此外，我们还确定了与年龄、性别和情绪表达相关的跨受试者的成像模式。
我们建立了数据融合技术可以识别与 OCD 表型相关的局部多模态神经影像学模式。结果为我们理解强迫行为和 OCD 诊断的神经生物学基础提供了信息。
