Transcript Effectiveness of Verb generation by noun and

Effectiveness of Verb Generation by Noun and
Picture naming of Thai Language Paradigm for
Functional MRI for Lateralization and
Localization of Language functional areas in
Thai Native Speaker
J.WONGSRIPUEMTET 1, P. SIWASATTAYANON 1S. MANOCHIOPINIG2, Y.
VICHIANIN3, O. CHAWALPARIT1
1 Department
of Radiology,2 Department of Physical Therapy, 3 Department of Radiation
Technique, Faculty of Technical Medicine, Mahidol University , Bangkok, Thailand.
Introduction

Brain activity study in clinical practice esp. neurosurgery

Goal of surgical treatment
 Curative
disease
 Preserve
normal brain function: eloquent areas such as
motor, sensory and language
 Need
physiologic brain study
Language function
 Target
1.
Where are the language areas ?(Broca’s & Wernicke’s
areas)

2.
of evaluation
Classic tools: EEG  high temporal resolution but less spatial
resolution
Which hemisphere is for language dominance (left VS
right)
 Right
handedness: 95% left / Left handedness: 70% left
 Gold
standard: Wada test
 Invasive,
discomfort and intervention risk
fMRI

fMRI is acceptable tools

Concordantly with Wada test

Adequate temporal resolution and high spatial resolution for evaluate brain
activity

Most of literatures use English paradigms

Thailand in 2011, Thai language paradigm (SiTP1)

Word generation

Verb generation

Naming picture

Sentence completion
Purpose

To localization the language association areas and
cerebral lateralization by using Verb generation (VG)
and Naming picture (NP) of Thai version of language
paradigm (SiTP1)
Material and methods
Material and methods

11 healthy volunteers

Thai native speaker

Right handedness by questionnaire of Edinburgh test

Normal visual and auditory tests

Normal language function: Battery of speechlanguage screening tests
Subjects

Exclude one volunteer
due to visual problem

10 volunteers:

5 male, 5 female

Age: 22-37 years old

Mean age 24.9 years
Number
Gender
Age
Education
Handedness
Language
Screening Test
(ALPS)
(Full Score = 40)
Communication
Checklist
Right
Auditory
Screening
Test
(Whispered
Voice Test)
Normal
1
Female
37
Grade 9
Normal (40)
Normal
2
Female
22
Postgraduate
student
Postgraduate
student
Postgraduate
student
Postgraduate
student
Postgraduate
student
Postgraduate
student
Postgraduate
student
Resident
Right
Normal
Normal
(39.5)
Normal
3
Male
23
Right
Normal
Normal
(40)
Normal
4
Female
22
Right
Normal
Normal
(40)
Normal
5
Female
22
Right
Normal
Normal
(39)
Normal
6
Male
25
Right
Normal
Normal (38.5)
Normal
7
Male
26
Right
Normal
Normal (39.5)
Normal
8
Male
24
Right
Normal
Normal
(40)
Normal
9
Female
29
Right
Normal
Normal
Postgraduate
student
Right
Normal
Normal
(40)
Normal (38.5)
10
Male
29
Normal
Language paradigms


Thai version of language paradigm (SiTP1)

Verb generation from noun (VG)

Naming picture (NP)
Block-paradigm

Silence response

Activate: 3 second for each noun or picture x 10 times
 Verb
generation by noun
 Naming

picture
Rest
 ###:
rest 30 second for block
Example of Verb generation by noun
Rest
Active
เสื ้อ
รถ
แก้ ว
Rest
Active ต้ นไม้ รถยนต์ แมว
###
ตู้ แว่นตา เก้ าอี ้ บ้ าน หมา ดอกไ ดินสอ
ม้
###
โต๊ ะ กระดาษ กางเกง ช้ อน รองเท้ า ไก่ ประตู
Please give verbs from given noun
Naming picture
Rest
###
Active
Rest
###
Active
Rest
Active
Rest
Please name the seen pictures
fMRI data acquisition
Parameter
fMRI
T1W anatomical image
Technique
SS-EPI
3D TFE
Voxel size
3x3x3 mm
1x1x1 mm
TR
3000 ms
7.6 ms
TE
35 ms
3.6 ms
FA
90
8
FOV
240x240 mm
230x290 mm
Slice thickness
5mm
1 mm
NEX
1
1
TFE factor
-
144
3T scanner(Achieva,
Philips Medical systems,
Best, NL)
Record Blood oxygen level-dependent (BOLD)
signal intensity
 Processing by SPM8

 Language

association areas: matching with Price
JC. Anatomical area
 Coordinate of Broca’s and Wernicke’s area: x-y-z
 Calculate lateralization index
Reviewed by two neuroradiologists and one fellow
neuroradiologist by consensus
Anatomical signal localization(Price 2010)
PMC
PP
vOCC/dOCC:
vOT/dOT:
aMTg/pMTg:
ITg:
T. pole:
HG:
STg:
aSTs/pSTs:
pPT:
PP:
vSMg/dSMg:
ANG:
pCen:
SMA:
SFg:
IFs:
MFg:
PMC:
pOr:
pTr:
vpOp/dpOp
ventral and dorsal occipital (approx. BA 18/19)
ventral and dorsal occipito-temporal (approx. BA37)
anterior and posterior middle temporal gyrus (approx. BA21)
inferior temporal gyrus (approx. BA20)
temporal pole (approximately BA38)
heschl’s gyrus
superior temporal gyrus (approx. BA 22)
anterior and posterior superior temporal sulcus
planum temporale, posterior to HG (approx. BA42)
posterior parietal gyrus
ventral and doral supramarginal gyrus (approx. BA40)
angular gyrus (approx. BA 39)
precentral gyrus (BA 6 anteriorly, BA 4 posteriorly)
supplementary motor cortex
superior frontal gyrus
inferior frontal sulcus
middle frontal gyrus (approx. BA 46)
premotor cortex
pars orbitalis (approx. BA 47)
pars triangularis (approx. BA 45)
ventral and dorsal pars opercularis (approx. BA 44)
Image analysis and preprocessing

Statistical Parametric Mapping software (SPM8, Wellcome Department
of Cognitive Neurology, London, UK) operation on LENUX system.

Preprocessing


Realignment

Coregistered with T1W image using rigid body transformation

Normalization to standard Montreal Neurological Institute (MNI) space

Smoothening by convolving with Gaussian kernel of 5 mm full-width at half
maximum (FWHM)
First level SPM t-contrast map for each subject at threshold of p<0.05
Hemispheric lateralization index

Add-on program in SMP8 is LI toolbox

LI between -1 to +1
 -1:
indicate right dominance
 +1:
 -0.2
indicate left dominance
to +0.2 : indicate bilateral or ambiguous
lateralization
Results
BOLD activity in each cerebral
hemispheres
BOLD signal intensity

9 in 10 - BOLD signals left > right in both paradigms

1 in 10 – no activity in both paradigms

Subject No. 4
BOLD SI in left hemisphere (number of
case in each cortex)
Naming picture
Verb generation
9
8
1
2
8
7
8
4
4
7
2
9
1
TOTAL = 10 CASE
2
1
8
7
5
3
2
2
5
5
6
9
8
7
5
6
9
4
5
6
2
2
3
6
2
2
2
8
3
3
2
2
6
7
8
BOLD SI in right hemisphere (number of
case in each cortex)
Naming picture
Verb generation
4
4
TOTAL = 10 CASE
1
3
3
1
1
3
2
2
2
1
1
3
2
5
2
1
2
7
2
1
2
1
2
4
7
1
7
7
7
Subject No. 7: BOLD-anatomical overlay
Verb generation
Naming picture
Coordinate
Broca’s area (9/10)
Verb generation
Case No. 4 demonstrated no
significant activity.
Naming picture
Case No. 4 demonstrated no significant activity.
Wernicke’s area
Verb generation
9/10 case
Case No. 5 showed no Wernicke’s area on NP
Case No. 8 showed two foci of Wernicke’s areas. (left ANG and left vSMG)
Naming picture
8/10 case
Lateralization index (LI)
Weighted mean of lateralization index
(LI) of whole brain on VG and NP


VG
 90%
- left hemispheric dominance
 10%
- ambiguous
NP
 40%
- left dominance
 50%
- ambiguous
 10%
- right dominance
Discussion
Thai language paradigm


Previous study (Wongsripuemtet et al)

Word generation Thai paradigm: effective for evaluation language function.

Other paradigms: also effective for localization activity and lateralization
Verb generation and Naming picture paradigms

Similar activating process

Picture paradigm: easier for illiterate in Thailand.

Goal of our study:
 Effectiveness
 Picture
of paradigm: Localization language areas and lateralization
paradigm could instead of verb generation paradigm
BOLD activity

fMRI cortical anatomy based on review paper of
Price CJ in 2010
 Coordinate

signal along xyz axis
Language association areas are beyond classical
Broca’s or Wernicke’s areas
 Concordance
 BOLD
with our study
activity predominantly on left hemisphere: left
frontal, left temporal, left parietal and both occipital
lobes.
Left frontal region

Left pars opercularis and pars triangularis
 Classical

Broca’s area
Left pre-SMA, SMA and PMC
 Mainly
associate with motor function
 Relate
to language process (Yang and Shu 2012)(Sun, Zheng et al.
2013)(Duffau, Capelle et al. 2003)
 Activated
on both paradigms, except PMC which not seen on NP
 Discordance
with previous studies  PMC was common during
naming process (Sun, Zheng et al. 2013)(Duffau, Capelle et al.
2003)
Left temporal and left parietal
regions

Left temporal (superior and middle) and left
parietal (angular gyrus and posterior parietal
lobe)
 Including
posterior superior temporal gyrus (classical
Wernicke’s area)
 Concordance
with previous studies (Binder, Frost et al.
1997)(Holland, Plante et al. 2001)(Rowan, Liegeois et
al. 2004)
Both occipital regions
 Activity
on both hemisphere and common
on both paradigms.
 Correspond
 No
with primary visual cortex
definite lateralization
 More activity on NP than VG
Visual word form area

Words or pictures activate
reading process
 Activity
 Visual
as figure
word form area
 Learning
to recognize word
forms (Cohen, Lehericy et al.
2002)
 However
our study was
represented overall activity
 cannot demonstrate the
sequential activity.
Coordinate

Broca’s area

VG and NP were similar results  most cases Broca’s area at pars
opercularis, esp. dorsal part


Activity at more or less than 20 mm above AC-PC line
Wernicke’s area

VG and NP showed variable areas at left superior temporal,
middle temporal, supramarginal gyrus, angular gyrus and
posterior parietal gyrus

Perisylvian region: concordant with previous studies
Subject No. 4
 No
activity during VG and NP
 Should
be from inadequate cooperation
during scanning
 Problem
of silent thinking response  cannot check
cooperation
 Some
studies using quite speaking response  but motion
artifact should be concerned

Study design and training volunteer before scanning
were essential.
Limitation

Paradigm design: block-paradigm
 Demonstrate
specific task
 Unable

only overall associated activity during
to evaluate connectivity
Small group of volunteers
 Inadequate
 Difference
for group analysis
age, education and disease
Conclusion
 Thai
language paradigm on verb
generation by noun and naming picture
were effective and useful for language
lateralization and localization language
association areas
Thank you