Multiple Weak Signal Parallel Acquisition Technology In FNIRS

Functional near-infrared spectroscopy imaging(fNIRS) is a non-invasive technology of brain activity detection with high temporal resolution which recently has been widespread in the field of medicine and clinical research.The main purpose of this paper is as the introduction of our research on multiple parallel weak signal acquisition technology embedded in the functional near-infrared spectroscopy imaging system. The method can be described as follows: based on the weak signal detection theory and the quadrature detector circuit, the 90 ° phase shifter circuit of DDS chip, supplemented by an additional high-precision analog-to-digital conversion module, ultimately forms the multiple weak parallel signal acquisition technology. It enabled simultaneous 32-channel signal acquisition of oxygen. In this paper, we adapted a continuous wave technology and superposed the low frequency modulated square wave(the maximum frequency is 40 KHz, the frequency spacing between channels 200 ~ 300Hz) on the intensity value of a stable light. We distinguished different BOLD signals in brain regions depended on various frequencies and, at last, achieved the goal of this article: synchronous 32 channels BOLD signal acquisition.The differences between the formal BOLD signal acquisition and ours are mainly from two aspects. 1, The formal oxygen signal acquisition technology based on continuous-wave technology focused on one specific time point with only one light channel. If there are multiple light sources, they need to be lighted one by one. On the contrary, the weak oxygen signals parallel acquisition techniques in this study can be set at the same time and all the light sources can be lighted without delay. Moreover, it is quite convenient to differentiate each brain region with the consideration of the square wave signal frequencies. 2, The formal oxygen signal acquisition technology generally use low-precision, high-speed analog-digital converter chip, so the oxygen signal acquisition was obtained in different time. Instead of that, using the frequency division multiplexing for BOLD signals, all the analog signals could be simultaneously converted to digital signals. To achieve this, in this paper, we use several low-speed high-precision analog-to-digital conversion chips conversion. Thus the BOLD signal can reach all channels with analog-to-digital conversion for the purpose of the real-time display and being stored in a PC for further signals processing.In the end of this paper, we inferred the specific parameters of the FNIRS subsystem model to explain its performance. And the Milk-ink experiment suggested its good quality in the practical application. Besides, the circuit board of 8-channle weak signal detection system has been detected.