Construction and validation of a database of head models for functional imaging of the neonatal brain

Abstract

The neonatal brain undergoes dramatic structural and functional changes over the last trimester of gestation. The accuracy of source localisation of brain activity recorded from the scalp therefore relies on accurate age‐specific head models. Although an age‐appropriate population‐level atlas could be used, detail is lost in the construction of such atlases, in particular with regard to the smoothing of the cortical surface, and so such a model is not representative of anatomy at an individual level. In this work, we describe the construction of a database of individual structural priors of the neonatal head using 215 individual‐level datasets at ages 29–44 weeks postmenstrual age from the Developing Human Connectome Project. We have validated a method to segment the extra‐cerebral tissue against manual segmentation. We have also conducted a leave‐one‐out analysis to quantify the expected spatial error incurred with regard to localising functional activation when using a best‐matching individual from the database in place of a subject‐specific model; the median error was calculated to be 8.3 mm (median absolute deviation 3.8 mm). The database can be applied for any functional neuroimaging modality which requires structural data whereby the physical parameters associated with that modality vary with tissue type and is freely available at http://www.ucl.ac.uk/dot-hub.

Longitudinal infant fNIRS channel-space analyses are robust to variability parameters at the group-level: An image reconstruction investigation

“…In addition, we had no access to any individual-level data from infants at the ages under investigation. This underlines the need for more publicly-available high-quality MRI data from infants in this age range, similar to the database of individual-level structural priors presented by Collins-Jones et al. (2020) .…”

“…For each age, the head model was warped by the group-average head measurements and the group-average array positioning data was used to register optode positions. The midpoint on the scalp surface between source and detector for each channel was projected onto the cortex in a method analogous to that which was demonstrated in ( Collins-Jones et al., 2020 ) which employs the Möller-Trumbore algorithm ( Möller and Trumbore, 1997 ; Mena-Chalco, 2019).…”

Group-level cortical functional connectivity patterns using fNIRS: assessing the effect of bilingualism in young infants

“…However, a recent work demonstrates that differences in head size and shape do not lead to substantial differences in channel positions and localization in infants (i.e., differences in head size make little difference over the range of head circumferences for a given age). 76 It is also challenging to ensure a consistent probe positioning when acquiring fNIRS data from infant subjects, as most methods for optode localization require participants to remain motionless for long periods of time, which is generally not suitable. Recently, photogrammetry methods that provide faster and easier optode registration procedures demonstrated their validity and reliability for probe position estimation.…”

The neural correlates of inhibitory control in 10-month-old infants: a functional near-infrared spectroscopy study

“…The inner skull border was delineated by the outside boundary of the cerebral tissue mask, while the scalp surface was defined using the Betsurf procedure (Jenkinson et al, 2005) where the group-level T1-weighted MRI template was used as an input. All voxels situated between the inner skull border and the scalp surface were assigned to be extra-cerebral tissue; this is a combined label for scalp and skull, and is commonly done in infant head models due to the difficulty in discerning these two tissues in infant MRI data (Brigadoi et al, 2019;Collins-Jones et al, 2021;Frijia et al, 2021). The resulting four-layer tissue mask (consisting of white matter, grey matter, cerebrospinal fluid and extra-cerebral tissue) was converted to a tetrahedral volume mesh and a grey matter surface mesh using the iso2mesh package ( (Fang & Boas, 2009), see iso2mesh.sourceforge.net).…”