[Paper Review] SQUID-based multichannel system for Magnetoencephalography

S. Rombetto, C. Granata+4 more|arXiv (Cornell University)|Oct 17, 2013

Atomic and Subatomic Physics Research3 citations

TL;DR

This paper presents a 163-channel SQUID-based magnetoencephalography (MEG) system developed at the Istituto di Cibernetica in Naples, featuring 154 axial gradiometer SQUIDs and 9 reference vector magnetometers in a helmet-shaped array. The system achieves a magnetic white noise level of approximately 5 fT/Hz¹², demonstrating stable, high-sensitivity performance suitable for clinical and routine neuroscience applications in a magnetically shielded room.

ABSTRACT

Here we present a multichannel system based on superconducting quantum interference devices (SQUIDs) for magnetoencephalography (MEG) measurements, developed and installed at Istituto di Cibernetica (ICIB) in Naples. This MEG system, consists of 163 full integrated SQUID magnetometers, 154 channels and 9 references, and has been designed to meet specifications concerning noise, dynamic range, slew rate and linearity through optimized design. The control electronics is located at room temperature and all the operations are performed inside a Magnetically Shielded Room (MSR). The system exhibits a magnetic white noise level of approximatively 5 fT/Hz1=2. This MEG system will be employed for both clinical and routine use. PACS numbers: 74.81.Fa, 85.25.Hv, 07.20.Mc, 85.25.Dq, 87.19.le, 87.85.Ng

Motivation & Objective

• To develop a multichannel SQUID-based MEG system with high sensitivity and stability for clinical and routine use in neuroscience research.
• To minimize sensor noise and cross-talk through optimized SQUID design, integrated feedback circuits, and compact dewar geometry.
• To ensure reliable performance in a real-world clinical setting by addressing thermal, magnetic, and mechanical noise sources.
• To achieve a magnetic field noise floor of approximately 5 fT/Hz¹², suitable for detecting weak biomagnetic signals from the brain.
• To validate system performance through characterization of the magnetically shielded room (MSR), thermal imaging, and seismic noise analysis.

Proposed method

• The system employs 163 full-integrated SQUID magnetometers, including 154 axial gradiometers and 9 vector magnetometers (3 orthogonal triplets), arranged in a helmet-shaped array.
• SQUID sensors are fabricated using trilayer superconducting technology with integrated flux transformers and on-chip Additional Positive Feedback (APF) and Flux-Locked-Loop (FLL) circuits to enhance sensitivity and stability.
• The dewar is constructed from fiberglass to minimize magnetic interference and reduce the distance between sensors and the scalp to 2 cm, with internal mylar layers and a foam mold to reduce thermal losses.
• All control electronics operate at room temperature within a Magnetically Shielded Room (MSR), which provides passive shielding with attenuation factors of up to -107 dB at 20 Hz in the y-direction.
• Thermal imaging and seismic noise analysis are used to monitor thermal gradients and mechanical vibrations, respectively, to identify and mitigate environmental noise sources.
• System performance is evaluated via spectral density measurements, MSR transfer function analysis, and noise floor characterization at 4.2 K.

Experimental results

Research questions

• RQ1 What is the achievable magnetic field noise floor of a multichannel SQUID-based MEG system designed for clinical use?
• RQ2 How does the integration of on-chip APF and FLL circuits affect SQUID stability and signal-to-noise ratio in a high-density array?
• RQ3 To what extent does the magnetically shielded room (MSR) attenuate external magnetic fields across the 0.01–20 Hz frequency range?
• RQ4 How do thermal gradients and mechanical vibrations in the laboratory environment impact SQUID performance, and what mitigation strategies are effective?
• RQ5 Can a compact, fiber-glass dewar with optimized thermal insulation maintain sufficient cryogenic performance for daily clinical operation?

Key findings

• The MEG system achieves a magnetic white noise level of approximately 5 fT/Hz¹², which is stable and suitable for high-fidelity biomagnetic signal detection.
• The MSR provides passive shielding with attenuation factors of -35.26 dB at 0.1 Hz (x-direction) and -34.56 dB at 1 Hz, increasing to -107.51 dB at 20 Hz in the y-direction.
• Thermal imaging revealed temperature variations within 1 °C across the dewar, primarily localized at joints and the helmet region, indicating effective thermal insulation.
• Seismic noise analysis identified significant low-frequency mechanical noise below 1 Hz, prompting plans for antivibration pads and active noise compensation.
• The system's SQUID sensors exhibit stable performance with a flux-capture area of 3 mm² and a flux-field conversion factor of 0.7 nT/Φ₀, enhancing sensitivity.
• The 74-liter dewar maintains liquid helium for approximately 7 days between refills, enabling continuous operation in a clinical setting.