Prefrontal Cortical Folding of the Preterm Brain: A Longitudinal Analysis of Preterm-Born Neonates - Longitudinal Registration and Shape Modeling - page 2

Of particular interest, the prefrontal cortex (PFC) is situated in the anterior
part of the frontal lobes of the brain, inferior to the motor and premotor areas. This
neocortical region is thought to play an important role in cognitive control, exec-
utive function and habituation [ 4 ]. Because of its anatomical connections with the
cortical and subcortical centres, important for movement control, the PFC plays a
role in coordinating motor function. Thus, accurate measurements in this region,
in particular its volume and shape change, might be predictive of early delays in
motor control. The development of the superior, middle and inferior gyri of the PFC
takes place mostly during the last trimester of pregnancy, later than the parietal
and occipital cortex, making it possible to study it using longitudinal data of infants
aged between 30 and 40 weeks estimated gestational age. The superior frontal gyrus
of the PFC becomes defined by 25 weeks of gestation [ 5 ]. The inferior frontal sulcus
is visible by 28 weeks gestational age (GA), followed by the delineation of the mid-
dle and inferior gyri [ 5 ]. All three main gyri shows secondary gyri at about 32 weeks
GA, while tertiary gyri are distinctive by 40 weeks GA [ 5 ]. Due to its development
timing, of the PFC may be affected by preterm birth.
Matching of cortical surfaces is a challenging process and most methods that
address this problem are are based on either optimising flows, such as LDDMM
[ 6 ], or on inflating surfaces to a common template which is usually a sphere, such
as FreeSurfer [ 7 ] and Spherical Demons [ 8 ]. However, both of these methods are
computationally expensive. Spectral graph methods offer a fast alternative to
matching shapes in the spectral domain [ 9 ].
In this paper we use Joint-Spectral Matching techniques [ 9 ] to measure the
longitudinal change in the cortical folding patterns of the same preterm-born
infants between 30 weeks equivalent gestational age (EGA) and term equivalent
age. Determining such correspondence yields information about the dynamics of
cortical folding, shape change, surface areas and volume growth rates of different
regions. This type of longitudinal correspondence will enable the understanding
of the development of the PFC during this crucial period, how it is affected by
preterm birth and how it might influence neurological outcome. Furthermore
this type of research might begin to illuminate the debate on the mechanical
role of tissue growth on the observed cortical folding pattern, information that
is only measurable in feral and neonatal cohorts of this type.

2 Methods

Subjects. Volumetric T 1 -weighted images were acquired for five infants (Mean
Gestational Age at Birth (GAB) of 26 . 0 0 . 9 weeks) acquired on a Philips

Achieva 3T MRI machine. T 1 -weighted data was acquired at a resolution of
0 . 82

±

0 . 5 mm at TR/TE =17/4.6 ms , acquisition duration 462 s .The
infants were scanned at first at around 30 weeks EGA (33w+1d, 31w+3d, 31w,
29w+6d, 31w+6d) and then at around 40 weeks EGA (40w+1d, 42w, 42w,
46w+2d, 40w+2d).

×

0 . 82

×

Infant Brain Segmentation. Brain masks and priors for automated segmen-
tation of the 40 week EGA scans were propagated from a publicly available