HARDWARE IMPROVEMENTS FOR DETECTING MODULATED NEAR-INFRARED LIGHT IN DIFFUSE OPTICAL MEASUREMENTS OF THE HUMAN BRAIN

Abstract

Functional near-infrared spectroscopy (fNIRS) is a non-invasive in vivo brain imaging method. fNIRS systems can be used to detect diseases that alter the hemodynamics of the brain, but they are also applicable to the study of hemodynamics in healthy brains to investigate, for example, how brain hemodynamics change in response to external stimuli. The work carried out in this thesis involved improving parts of the light detection hardware that forms the core of a frequency-domain/spatially-resolved fNIRS system. Different narrow band-pass filter configurations were analyzed to determine the best option for a system of that type. Among the various alternatives, multifeedback filters and resonator filters were simulated and compared. Finally, two different resonator filter sets were physically implemented using printed circuit technology. The second part of the thesis describes the manufacture of six low-noise, highsensitivity NIR light detectors designed to extend the light detection capabilities of the system. All these detectors were implemented on printed circuit boards. After implementation, the circuits were tested separately and in combination with other parts of the system, achieving good results in both cases. The most significant result was the detection of blood flow pulsations from the finger and forehead of a subject using the designed light detectors in combination with the designed filters and a lock-in amplifier. This result shows that the circuits are fully functional and can be used to expand the capabilities of the fNIRS device