SQUID based multichannel system for brain functional imaging

A multichannel system for brain imaging containing 163 SQUID magnetometers arranged in a helmet shaped multisensorial array has been developed. To this aim, a previous investigation of a several SQUID configurations has been performed in order to choose a SQUID sensor having best performance for brain imaging on the basis of system working conditions. In particular, magnetometer and planar gradiometer have been designed, fabricated and characterized. Furthermore, a small magnetometer has been also taken into account. Since it has been decided to work in a magnetically shielding room a SQUID magnetometer has been chosen in order to guarantees high magnetic field sensitivity. These SQUID magnetometers are based on an integrated Ketchen design including a superconducting flux transformer consisting of a pickup coil and a multiturn input coil inductively coupled to the SQUID loop in a washer shape. The circuits for Flux Locked loop (FLL) operations and for Additional Positive Feedback including a thin film resistor network for gain adjusting, are integrated on the same chip containing the SQUID magnetometer. A magnetic field spectral noise as low as 1.8 fT/Hz measured in FLL operation, ensures the sensor capability to detect the tiny magnetic filed arising from brain activity. More than 200 SQUID magnetometers have been fabricated and characterized selecting the sensors with best performance. 154 measurement sensors are arranged in a SQUID multisensorial array properly designed and customized realized. Further 9 channels are located far from the scalp on three bakelite towers in order to realize via software a gradiometer to background noise rejection. The system properties have been investigated including the estimation of background noise, the shielding factor of the magnetically shielding room and the Dewar performance. Finally, preliminary measurements have been successfully performed. In particular, some brain activities such as the alpha rhythm, a spontaneous activity of a human having closed eyes and the evoked activity concerning the tapping of left and right forefinger have been analyzed. The same activities have been co-registered using the 32-channels EEG to measure bioelectric activity. The good agreement between EEG and MEG data indicates that the system operates properly.