PDF File - Mahidol University

584
Abstracts
FP-M-002
3-D MRI volume of frontal lobe in autistic children
Y. Toda, K. Mori, K. Tao, K. Inoue, M. Miyazaki, T. Saijo,
T. Hashimoto, Y. Kuroda
Department of Pediatrics, School of Medicine, University of
Tokushima, Tokushima, Japan
Purpose: We measured volume of the whole brain and
frontal lobe using 3-D MR image and coherence with laterality
and DQ examined volume of frontal lobe. Object/
method: Object is totaled 35 examples of 19 autistic children
(3–5 years old, DQ 50–100) and 16 healthy contrasts (3–5
years old) visiting our hospital. We got informed consent
from family with both group about MRI examination and
measured horizontal image with 3-D SPGR method. We did
reconstitution using image analysis software ‘Scion Image’
and got the volume of the whole brain and frontal lobe and
surveyed volume of right and left each about frontal lobe.
We used Enjyoujishiki developmental test about DQ.
Result: We did not recognize significant difference between
autism group and healthy control group about the volume of
the whole brain and frontal lobe. We compared even rate for
volume of the whole brain (%) about frontal lobe, but did
not still recognize significant difference. The right side was
big in significance compared with the left with autism group
in comparison of right and left frontal lobe volume. We did
not recognize correlation during brain and frontal lobe
volume about DQ. Conclusion: In volume of frontal lobe,
the right side is big in significance compared with the left
with autism group and coherence with frontal lobe dysfunction
was suggested.
FP-M-003
Clinical and magnetic resonance image analysis of 36
cases of periventricular leukomalacia in children
X.-D. Yuan
First People’s Hospital of Shangqiu City, Shangqiu, Henan,
China
Objective: To investigate the clinical and MRI characteristics
of PVL in children. Method: To analysis the clinical
data of PVL in 36 children. Result: (1) Gestational age:
between 30 , 37 weeks in 22 cases, 38 , 42 weeks in 14
cases. (2) Clinical manifestation: (a) paralysis in 36 cases;
(b) cortical blindness in two cases; (c) mental retardation in
nine cases; and (d) epilepsy in ten cases. (3) MRI manifestation:
(a) foci form: patchy areas in 16 cases, punctate areas
in 13 cases, streaks in seven cases; (b) section: symmetric
signal abnormalities were observed in the periventricular
and the subcortical white matter; (c) the signal of the
focus: hyperintense/intense on T1 weighted and hyperintense
on T2 weighted; and (d) other: bilateral ventricular
bodies dilatation in eight cases. Trigon dilatation in five
cases. Conclusion: PVL is a main cause of cerebral palsy,
especially of spastic paralysis, MRI is a main way to diagnose
PVL in early infants.
FP-M-004
Automated histogram-based brain segmentation and
volume measurement in T1-weighted 3-D MR head
images
J.Z. Liu, Z.Y. Shan, G.H. Yue
Department of Biomedical Engineering, Cleveland Clinic
Foundation, Cleveland, OH, USA
Information on human brain volume changes may help
better understand structural and functional development and
adaptations of the brain. Current semi-/automated MRbased
brain segmentation and volume measurement methods
are complex and not sufficiently accurate. We have
developed a simpler, more accurate automated algorithm
for whole-brain segmentation and volume measurement in
T 1 -weighted 3-D MR images. This histogram-based brain
segmentation (HBRS) algorithm is based on histogram
analysis and simple morphological operations. It includes
three major steps: (1) foreground/background thresholding;
(2) disconnection of brain from skull and other head tissues;
and (3) removal of residual fragments (cerebrospinal fluid,
sinus, dura and marrow). Brain volume was measured by
counting the number of brain voxels. Accuracy was determined
by comparing brain volume rendered by applying the
algorithm to computer-simulated MR data with the volume
on which the simulation was based; average error was
1.48%. Reproducibility of brain volume measurements
was assessed by comparing data from two sessions (two
trials in each session) with human volunteers. Intra-session
variability of brain volumes for sessions 1 and 2 was
0.55 ^ 0.56 and 0.74 ^ 0.56%, respectively; mean difference
between the two sessions was 0.60 ^ 0.46%. These
results show that the HBRS algorithm is a simple, fast,
and accurate method for human brain volume assessment
with high reproducibility. This algorithm may find various
applications in research and clinical investigations regarding
cortical development and child neurology.
FP-M-005
Diffusion weighted MRI for early diagnosis of acute
neurological disease in the neonate
W. Brussel, I.M. van Beynum, M.C. Molenschot, V.A. Bok
Rijnstate Hospital, TA Arnhem, Netherlands
Diffusion weighted-MRI (DW-MRI) uses water movement
and transport of water in biological tissues as a source
of contrast. Therefore DW-MRI can be used to obtain information
about molecular displacement over distances
comparable to cell’s dimensions and, consequently, about
geometry and spatial organization of tissue compartments.
Functional insight concerning exchanges between these
compartments in various disease states can be gathered. In
the presence of magnetic field gradients, protons carried by
free moving water molecules, undergo a phase shift of their