Active Noise Cancellation

* Magnetic Resonance Imaging (MRI) has outstanding contribution to medical imaging. But MRI will produce the noise over 100dB. This volume, that people can only withstand 30 minutes of scan time, limits image quality. The study is to discuss how to reduce the noise.

* There are many papers to discuss and resolve the noise problem of the magnetic resonance images. Like to improve the magnetic field coil, set the vacuum layer isolated from the media and other methods can effectively rectify the noise problem, but above ways are too expensive, and cannot be used in the existing medical equipment.

* The most traditional way is to use the passive materials such as earplugs and earmuffs to achieve the anti-noise effect, but the long-wavelength low-frequency noise around these materials still need to be improved. More practical method is the active noise elimination. Use the horn to create a backward wave, cancellation through a wave of positive and negative, to achieve the anti-noise effect. Because of the delay time of the instrument, long-wavelength low-frequency noise is easier to be cancelled, just to make up for the shortcomings of the passive material. With the help of active and passive methods of noise elimination, we aim to achieve the best effect on anti-noise. Although there are active anti-noise methods, we use the widely applied and more stable FxLMS adaptive algorithm, using 100-point finite impulse response filter to implement this algorithm.

* The framework of this experiment is an active noise suppression system, using one of the fastest digital signal processor, with the signal amplifier produced in experimental demand, and a good set of headset and microphone to achieve the purpose of noise reduction. In this system, 1000Hz single frequency noise can reach 45 dB noise reduction and 7.5 dB noise reduction in Echo-planar Imaging.

* We look forward to better and more stable results in the future. With the help of well-equipped non-ferromagnetic material, this system can actually be used in MRI system and proceed the auditory and linguistic MRI study. Keyword: active noise suppression, magnetic resonance imaging

Effects on MRI Noise

*Uncomfortableness for participants

*Interference in the communication between participants and MR operator

*Specifically for functional MRI

  1. Unwanted activation at auditory cortex
  2. Interference in auditory stimulation

Feedforward ANC system

*Reference mic receives EPI noise only. X(n)

*Headsets transmit the reverse wave. Y(n)

*Error mic receives the EPI noise and the reverse wave. e(n)

System flow

* system setup and geometry graph

Filtered-X-LMS

*e(n)=d(n)-y’(n)

*y’(n)=y(n)*s(n)

*w(n+1)=w(n)+μx(n)s(n)e(n)

μ is a convergence factor (or step size)

Result

*FxLMS gets 12.4dB noise reduction for EPI in simulation ,7.7dB in real time experiment.

*Frequency response is more important than time delay in this system.

*Hardware improvement may enhance the result to ideal situation.

*This real-time system can be safely exposed to MRI environment.

Our Journal and Conference papers

  1. C. K. Chen, C. Tzi-Dar, and C. Jyh-Horng, “Active cancellation system of acoustic noise in MR imaging,” Biomedical Engineering, IEEE Transactions on, vol. 46, pp. 186-191, 1999.
  2. C. Kuan-Hung, C. Tzi-Dar, L. Ching-Po, C. K. Chen, and C. Jyh-Horng, “An Active Noise Cancellation System for fMRI,” in Noninvasive Functional Source Imaging of the Brain and Heart and the International Conference on Functional Biomedical Imaging, 2007. NFSI-ICFBI 2007. Joint Meeting of the 6th International Symposium on, 2007, pp. 265-267.
 
pub/para/anc/start.txt · Last modified: 2012/10/25 17:02 by daniel     Back to top