Clinical Applications of fMRI

1
Clinical Applications of fMRI

• Mapping sensorimotor cortex and SMA
• Mapping Brocas and Wernickes areas
• Determination of hemispheric dominance for
  language
• Mapping visual cortex
• Mapping auditory cortex
• Studying the organization of cognitive functions
  in patients with psychiatric and
  neurodegenerative disorders

2
Mapping language functions with fMRI

• Victor Haughton MD
• Robert Dempsey MD
• Department of Neurosurgery
• M Elizabeth Meyerand
• Chad Moritz
• Departments of Neurosurgery, Radiology, Medical
  Physics

3
Mapping language functions with fMRI

• Language areas in the brain
• Paradigms for mapping language areas
• Clinical applications
• Localize language area
• Hemispheric dominance for language
• Study language organization in language disorders
• Evaluation of precision and accuracy of language
  activation

4
Mapping language functions with fMRI

• Frontal lobe and posterior language areas
• Variety of paradigms for mapping language areas
• Comparison of results between paradigms
• Clinical applications
• Precision of language functions

5
Language areas

• Brocas
• Brodmann area 44, 45, 6

• Wernickes
• Brodmann area 22

6
Language Paradigms

• Word generation from auditory cues
• Letters, Classification, Antonyms, Verbs etc
• Word generation from visual cues
• Rhyming
• Listening to text
• Listening to text/complex sounds
• Reading text
• Foreign language
• Sign language

7
A
B
C

• word generation from antonyms
• word generation from a given alphabet letter
• text reading

8
A
language comprehension task visually presented
9
A
B
Text reading receptive language paradigm in two
patients. Activation is evident in the left
inferior parietal lobe. Both patients were
studied prior to resection of cavernous
hemangiomas, not shown in the images.
10
A
B

• Word generation expressive language paradigms in
  a patient with left frontal cavernous angioma
• Word generation in a patient with left frontal
  tumor.

11
Results in 31 patients
12
Clinical indications for fMRI

• Determination of hemispheric dominance
• Localization of Brocas and Wernickes areas
• Localization of Supplementary Motor Area (SMA)
• Study organization of language functions in
  patients with psychiatric disorders

13
RD, Temporal-frontal glioma

• Text listening
• task
• Relationship of
• activation to
• lesion

14
fMRI mapping of word generation in a 42 year old
left handed dentist to determine hemispheric
dominance
15
fMRI of word generation in a left handed patient
to determine hemispoheric dominance
16
SB, L temporal lobe insular mass

• Activation from a
• word generation
• task
• L hemispheric
• dominance for
• language

17
JQ, L temporal lobe cyst seizures, evaluated
for possible surgical excision of the
epileptogenic focus

• Relationship of
• cyst and activation
• from a text listening
• task

18
SS, large L frontal tumor

• Activation from
• word generation
• task
• Mixed hemi-
• spheric dominance
• for language

19
SS, large L frontal tumor

• Activation from
• word generation
• task
• Mixed hemi-
• spheric dominance
• for language

20
RD, Temporal-frontal glioma

• Silent word
• generation
• task
• Relationship of
• activation to
• lesion

21
JA, 51 year old man with a frontal glioma. fMRI
to determine relationship of tumor to SMA
Alphabet Letter Generation
Motor Task
22
Precision of language activation paradigms
23
Precision of language activation

• Location of activation similar for various tasks
• Location of fMRI activation similar to
  localization of function by ISM
• Voxel precision of language tasks similar to that
  for for motor and visual tasks
• Laterality index similar for different tasks

24
Letter Word Generation First Iteration
Letter Word Generation Second Iteration
25
Letter Word Generation First Iteration
Letter Word Generation Second Iteration
26
First iteration
Intersect map
Second iteration
Letter Word Generation Test-Retest precision
27
Test-retest precision of fMRI of letter word
generation language task

• Concurrence ratios from 17 to 57
• Average concurrence ratio (voxel precision) 37)

28
Precision of fMRIDifferences between tasks
29
Letter Word Generation
Antonym Word Generation
30
Letter Word Generation
Antonym Word Generation
31
Category Word Generation
Letter Word Generation
32
Letter Word Generation
Category Word Generation
Intersect Map
Letter Word Generation/ Category Word
Generation Comparison
33
Letter Word Generation
Antonym Word Generation
Category Word Generation
34

• Activation map for AWG
• Activation map for TR
• Intersect image

35
TR
Intersect Image
AWG
36
AWG
Intersect Map
LWG
37
Comparison of results between paradigms Brocas

• AWG and TR correlate
• Both agree on right dominance and left dominance
• Less agreement in the mixed dominance cases

38
Comparison of results between paradigms
Wernickes

• More examples of mixed and right dominance
• AWG and TR correlate
• Both agree on right dominance and left dominance
• Less agreement in the mixed dominance cases

39

• Precision of left and right hemisphere language
  regions
• Average concurrence ratios from 25 to 45 for
  language areas
• Compares with average concurrence ratios of 50
  for motor and visual systems
• Ratios highest for LWG/AWG

40
Accuracy of fMRI Agreement with Intraoperative
stimulation mapping
41
B
C
A
ESM
fMRI
ESM Vs. fMRI
42
Agreement of fMRI and Intraoperative stimulation
mapping

• Seven patients with both ISM and fMRI
• Activation in same area as intraoperative
  localization in every case
• Activation in additional areas in 4 cases
• fMRI sensitivity high
• fMRI specificity low

43
Summary

• Frontal lobe and posterior language areas
  activated by expressive and receptive language
  tasks
• Similar results from multiple paradigms
• High intrasubject precision between language
  paradigms
• Robust clinical tool
• Map Brocas area, Wernickes area
• Determine hemispheric dominance for language