Utilizing multivariate pattern analysis to uncover the neurobiological underpinnings of communication disorders in...

Bilateral spastic cerebral palsy (BSCP) is frequently associated with communication impairments, with magnetic resonance imaging (MRI) revealing morphological and connectivity changes. We develop an individual diagnostic model for communication impairment using multivariate lesion and connectome-based approaches in BSCP children.

A total of 28 children aged from 4 to 16 years diagnosed with BSCP and 31 matched typically developing children were recruited. All children received three-dimensional T1-weighted imaging (3D-T₁WI) and diffusion tensor imaging (DTI) scans. Communication impairments were evaluated using the verbal comprehension index (VCI) and Communication Function Classification System (CFCS). The correlation between the MRI morphological and diffusion parameters and VCI and CFCS was analyzed, and support vector classification (SVC) algorithm was used to construct a diagnostic model of communication impairment in children with BSCP.

Children with BSCP showed reduced mean cortical surface areas and gray matter volumes mainly in the frontal and temporal lobes, along with widespread decreases in white matter connectivity [false discovery rate (FDR) correction, P<0.05]. The morphological alterations and white matter fiber of bilateral frontal lobes, sensory motor areas, and part of temporal occipital lobes were associated with communication impairment of BSCP children (P<0.05). The combined SVC model, based on gray matter morphology and white matter fiber index, can be utilized for diagnosing communication impairments in children with BSCP, achieving an accuracy of 80.77% and an area under the curve of 0.88. The cortex features that distinguished communication impairment in children with BSCP were predominantly found in the bilateral middle frontal gyrus, left middle temporal gyrus, and the white matter fiber features were primarily located within and between the bilateral frontal lobes, sensorimotor areas, and partial temporo-occipital lobes.

The brain morphological and white matter connectivity changes are associated with communication impairment in BSCP children. The multi-parametric MRI can be used to establish an individualized diagnosis model of communication impairment in children with BSCP.