RSNA 2003 Scientific Papers > fMRI of Auditory Activation from Language Stimulation ...
  Scientific Papers
  SESSION: Pediatric (Pediatric Neuroradiology: MR Imaging, Developmental)

fMRI of Auditory Activation from Language Stimulation in Neonates

  DATE: Wednesday, December 03 2003
  START TIME: 10:30 AM
  END TIME: 10:37 AM
  LOCATION: Room N229
  CODE: K15-987

Shantanu Sinha PhD
Los Angeles , CA
Susan Bookheimer PhD
John Grinstead
Meena Garg
Lina Badr PhD

Magnetic Resonance Imaging
Brain, function
Magnetic resonance (MR), experimental studies
Magnetic resonance (MR), functional imaging
Virtual Presentation

Purpose: The overall objective of this study was to monitor in a longitudinal manner, at several time points over a 24 month period, the effect of home-based cognitive and sensori-motor intervention on the development of infants who have suffered brain injury, using amongst others, fMRI in an effort to quantify the efficacy of the interventional program towards enhancing neuroplasticity. We report herein our observations using fMRI towards elucidating the question of the development of left hemisphere laterality for language in neonates.

Methods and Materials: Till this point, a total of 42 subjects, below the age of 6 months, with documented brain asphyxia, seizure or prolonged acidosis, were recruited. While no sedation was utilized, the babies were well fed, 30~40 mnts. before MR scan, put in a "swaddle bag", and put to sleep, with the lights dimmed. The entire baby was put inside the head coil of a 1.5T, LX GE scanner with EchoPlanar capabilities. Two sets of ear muffs were utilized to insulate for the EPI sounds and allow transmission of auditory stimuli. Vital signs were monitored. A 3D stack of axial high resolution images were first collected for registration and morphology purposes, followed by a 2 mm stack of T2-wtd slices for pathology. Gradient Echo EPI, (18 cm FOV, 10 slices, 5 mm thick, TE/TR: 60/2500ms) was used for fMRI. The auditory stimuli paradigm, recorded on an audio tape and played through the head-phones, consisted of two sets of three blocks, each block of 10 sec of scanner noise, Nonsense speech and "Motherese Speech". In each block of 20 sec, 8 phases of each slice were acquired. The results were analyzed using SPM99. Images were realigned, smoothed, normalized and convoluted with a slow hemodynamic blood-flow response. Temporal pixel-wise cross-correlation of signal intensity during speech and rest were calculated, with a cutoff value of 0.30 for Pearson's r statistics for 6 or more voxels (P<0.01).

Results: Strong activity in Broca's and Wernicke's area was detected during speech. While bilateral activity was observed in primary auditory cortex, left hemisphere bias was observed. In posterior superior temporal lobe, activity was found in the left hemisphere. Inferior frontal gyrus activity was minimal while Wernicke's area had unilateral activity.

Conclusion: These preliminary observations indicate that infants show lateralized activation in language areas in response to speech. Evidence for strong laterality in Wernicke's area suggests that left hemisphere dominance is established at birth.




Questions about this event email: