fNIRS

Optical & ultrasound (in vivo) · reads

fNIRS is a noninvasive optical brain-imaging method in which sensors on the scalp shine near-infrared light (650 to 950 nm) into the head and detect the light scattered back at several wavelengths. From how much light is absorbed across wavelengths and over time, it estimates changes in oxygenated and deoxygenated hemoglobin in the outer brain, an indirect, blood-flow-based sign of brain activity. Continuous-wave systems give only changes from a baseline, and blood flow in the scalp can contaminate the signal and be mistaken for brain activity.

Spatial resolution
8.2 mm (point spread)
Temporal resolution
20 ms (sample interval)
Field of view
51–112 cm³ (sensing envelope)
Reads
oxyhemoglobin (HbO2) concentration change; deoxyhemoglobin (HbR) concentration change
Destructive?
non-destructive
Prerequisites
other: Fit/place an fNIRS optode array or cap on the scalp; no surgical or genetic preparation is required.
Theoretical spatial resolution
8.2 mm (point spread)
Theoretical field of view
444–660 cm³ (sensing envelope)

Description

Functional near-infrared spectroscopy (fNIRS) is noninvasive optical neuroimaging. Light sources and detectors (optodes) on the scalp send near-infrared light of 650 to 950 nm into the head and record the intensity of the light scattered back at several wavelengths. Oxyhemoglobin (HbO2) and deoxyhemoglobin (HbR) absorb these wavelengths differently, so the method uses the change in attenuation across wavelengths and over time to estimate task-linked changes in HbO2 and HbR concentration in the cortex under the optodes. The result is a hemodynamic brain signal. The hemoglobin changes are estimated from the attenuation; they are not measured directly. Continuous-wave systems, the most widely deployed kind, give only changes relative to baseline, because they cannot separate absorption from scattering. Time-domain and frequency-domain systems separate the two and recover absolute concentrations, and time-domain systems also resolve depth and path length; both kinds are less available. The recorded signal includes blood perfusion and oxygenation in extracerebral tissue, especially the scalp, as well as in the cortex, and this scalp contribution can be mistaken for brain activity.

Download the dataset (JSON, described by its JSON Schema).

Loading…