A Neuronal Signal Detector for Biologically Generated Magnetic Fields

This paper presents a neuronal signal detector for biologically generated magnetic fields. The system includes a hardware section implemented with discrete electronics, which has an ultralow-noise dc or dc+ac current source for magnetoresistive sensor biasing, and signal amplification and filtering, and a software interface that allows signal demodulation, visualization, and digital postprocessing. Compared with the previous measurement setup, the results show that, for the same bandwidth, the proposed instrumentation system has approximately 50 times better noise performance, making the sensor noise the dominant noise source. The system is able to record the magnetic field generated by ionic currents from action potentials of in vitro experiments with mice brain slices. In addition, to obtain an increased spatial resolution, by scaling the number of sensors that can be read, and to enhance the system immunity to external interferences, two integrated circuits with an ultralow-noise current source for MR biasing and a low-noise variable gain amplifier were developed and are also presented.