A high-sensitivity magnetocardiography system with a divided gradiometer array inside alow boil-off Dewar

Abstract

We fabricated a low-noise 64-channel first-order axial gradiometer system for measuringmagnetocardiography (MCG) signals. The key technical features of the system are thecompact structure of the gradiometer, division of the sensor array plate, direct mounting ofthe sensor plates into the Dewar bottom, reduced neck diameter of the liquid He Dewar,and compact readout electronics. To make the refill interval of liquid He longer, thedistance between the compensation coil of the gradiometer and the input coil pads of thesuperconducting quantum interference device (SQUID) was reduced to 20 mm. By usingdirect ultrasonic bonding of Nb wires between the pickup coil wires and input coil pads,the superconductive connection structure became simple. The baseline of thefirst-order gradiometer is 70 mm, a little longer than for typical conventional axialgradiometers, to provide a larger signal amplitude for deep sources. The 64-channelgradiometer array consists of four blocks, and each block is fixed separately onto thebottom of the Dewar. The neck diameter of the He Dewar (192 mm) is smaller thanthe bottom diameter (280 mm) in which the gradiometers are distributed. Theaverage boil-off rate of the Dewar is 3 l per day when the 64-channel system is inoperation every day. Double relaxation oscillation SQUIDs (DROSs) having largeflux-to-voltage transfer coefficients were used to operate SQUIDs via compactelectronics. The magnetically shielded room (MSR) has a wall thickness of 80 mm, andconsists of two layers of permalloy and one layer of aluminum. When the 64-channelsystem was installed inside the MSR, the field noise level of the system was about3.5 fTrms Hz−1/2 at 100 Hz. MCG measurements with high signal quality were done successfullyusing the developed system. In addition to the parameter analysis method, wedeveloped software for the three-dimensional imaging of the myocardial currenton a realistic image of the heart based on the anatomical image of the torso.