Statistical parametric mapping: assessment of application in children.

Area of Science:

• Neuroimaging
• Pediatric Neurology
• Medical Imaging Analysis

Background:

• Statistical Parametric Mapping (SPM) is crucial for comparing functional imaging data across patient groups.
• Its application in pediatric studies is limited due to a lack of validation for spatial normalization in children of varying ages.
• This study addresses the feasibility of using SPM for pediatric functional neuroimaging.

Purpose of the Study:

• To evaluate the efficacy of spatial normalization of FDG PET images in children using the SPM96 technique with an adult template.
• To assess the impact of age on the accuracy of spatial normalization in pediatric brains.
• To determine the reliability of SPM analysis for identifying hypometabolism in pediatric epilepsy.

Main Methods:

• Applied PET-derived transformation parameters to coregistered MRI images from 13 children (2-14 years) with epilepsy.
• Compared contours of spatially normalized MRI images from children with those from 17 adult controls.
• Performed separate SPM analyses for children under 6 and those aged 6-14.

Main Results:

• Spatially normalized pediatric MRI contours showed greater variability than adult controls, with age-dependent deviations.
• SPM analysis in children <6 years revealed significant artifacts, precluding reliable application.
• SPM analysis in children 6-14 years showed no artifacts, identifying hypometabolism consistent with epileptic foci and similar glucose utilization patterns to adults.

Conclusions:

• Spatial normalization to adult templates is not recommended for children under 6 years due to significant errors.
• SPM analysis is viable for children aged 6-14, providing insights into brain metabolism in pediatric epilepsy.
• Children over 6 demonstrate similar glucose utilization patterns to adults, challenging previous findings of significant pubertal changes.