© 2000 - 2011 LIN - Leibniz Institut für Neurobiologie Magdeburg

LIN: Forschungsabteilungen > Akkustik, Lernen, Sprache > Unterpunkt Ebene 3 > Unterpunkt Ebene 4

Titel: LIN Layout

Home switch to german Print Search:
Facebook YouTube
Staff Links Sitemap

 Methods

 Functional magnetic resonance imaging (fMRI)

Low-noise fMRI

It has long been shown that the gradient noise generated by MRI scanners (which may exceed 120 dB using Echo Planar Imaging at 3 Tesla) leads to the activation especially of primary but also of secondary auditory cortex areas. To minimize this confounding effect, we use a modified FLASH sequence which offers the possibility of slowing down the gradient switching without affecting the image quality. This reduced the noise level by > 30 dB below 500 Hz. The headphone system (Fig. 1) and a foam pillow gave > 20 dB suppression of background noise for frequencies above 0.5 kHz and more than 30 dB suppression at 2 kHz. Furthermore, the scanner room is lined with an acoustic wall to suppress reverberation (Fig. 2). All these measures add up to a “low noise” imaging protocol with a noise peak level of 48 to 54 dB SPL at the subject’s ear depending on the gradient ramp time.

Gradient noise of scanner

Here you can hear four different types of gradient noise generated by our 3 Tesla scanner using different imaging schemes.

  1. EPI (Echo Planar Imaging) This sequence is not in use in our lab for auditoryMDEFT (Modified Driven Equilibrium Fourier Tomography) is used to obtain T1-weighted images at high magnetic field strengths. experiments.
  2. MDEFT (Modified Driven Equilibrium Fourier Tomography) wird verwendet, um T1 gewichtete anatomische Details bei hoher Feldstärke abzubilden.
  3. GEFI 250 (also called FLASH-sequence with 250 µs gradient ramp time). Though suited for fMRI, the short rise time causes still too much noise for auditory studies.
  4. GEFI 2500 (with 2500 µs gradient ramp time, 54 dB SPL). This gradient echo sequence is well suited for auditory fMRI. For studies in which the scanner noise is extremely critical we use GEFI 6000 with 48 dB SPL at the expense of imaging time.

Example WAV-File

Fig. 1: Capsules from commercially available headphones were modified by removing their magnets and integrating them into ear muffs with liquid-filled rims (Baumgart et al. 1998).

Fig. 2: Scanner room with acoustic wall to suppress reverberation.

last update: 2006-01-05 report a bug print this page