Real-Time Compensation of Power Frequency Interference and Eddy Current Magnetic Field for SQUID-Based Ultralow Field Magnetic Resonance Imaging

Quan Tao,Shaojie Liu,Mengmeng Yu,Yongqiang Li,Yi Xiao,Hui Dong
DOI: https://doi.org/10.1109/tasc.2022.3198740
IF: 1.9489
2022-01-01
IEEE Transactions on Applied Superconductivity
Abstract:In ultralow field magnetic resonance imaging (ULF MRI), extremely sensitive superconducting quantum interference devices (SQUIDs) and prepolarization field ( ${\mathrm{B}}_{\mathrm{P}}$ ) are used to enhance the signal-to-noise ratio (SNR). By placing the ULF MRI system in a conductively shielded room (CSR), external electromagnetic interference at frequencies in the kHz range, which is the typical ULF MRI imaging frequency range, can be effectively suppressed. However, CSR has a limited shielding effect on electromagnetic interference below 1 kHz, which are mainly the power frequency interference and its harmonics. In addition, eddy current is induced by the quick turn- off of ${\mathrm{B}}_{\mathrm{P}}$ current in the metal plates of CSR, which in turn induces an exponentially decaying magnetic field B EC in the imaging region. Power frequency interference and B EC dominate the voltage output of SQUID readout electronics and even exceed its output range. Thus, the gain of the SQUID readout electronics has to be decreased by reducing its feedback resistance. In this case, the analog-to-digital converter noise may start to contribute to the noise of ULF MRI. In this article, we describe a real-time compensation technique to suppress the large variation of the signal channel output due to power frequency interference and B EC . The experimental results demonstrate that this technique can significantly reduce the maximal variation range of the signal channel output by 79.2% and improve the SNR of ULF MRI images by 4.2 times. With this technique, we may use SQUID detectors with larger imaging areas and/or stronger ${\mathrm{B}}_{\mathrm{P}}$ fields to further improve SNR without the reduction of the gain.
What problem does this paper attempt to address?