A review on continuous wave functional near-infrared spectroscopy and imaging instrumentation and methodology

We report development, testing, and in vivo characterization of a multichannel optical probe for continuous wave (CW) functional near-infrared spectroscopy (fNIRS) that relies on silicon photomultipliers (SiPMs) detectors. SiPMs are cheap, low voltage, and robust semiconductor light detectors with performances analogous to photomultiplier tubes (PMTs). In contrast with PMTs, SiPMs allow direct contact with the head and transfer of the analog signals through thin cables greatly increasing the system flexibility avoiding optical fibers. The coupling of SiPMs and light-emitting diodes (LEDs) made the optical probe lightweight and robust against motion artifacts. After characterization of SiPM performances, which was proven to provide a noise equivalent power below 3fW, the apparatus was compared through an in vivo experiment to a commercial system relying on laser diodes, PMTs, and optical fibers for light probing and detection. The optical probes were located over the primary sensorimotor cortex and the similarities between the hemodynamic responses to the contralateral motor task were assessed. When compared to other state-of-the-art wearable fNIRS systems, where photodiode detectors are employed, the single photon sensitivity and dynamic range of SiPMs can fully exploit the long and variable interoptode distances needed for correct estimation of brain hemodynamics using CW-fNIRS.

Although functional near-infrared spectroscopy (fNIRS) has been developing as a useful tool for monitoring functional brain activity since the early 1990s, the quantification of hemoglobin concentration changes is still controversial and there are few detailed reports especially for continuous-wave (CW) instruments. By means of a two-layer model experiment mimicking hemodynamic changes in brain and mathematical analysis based on the modified Beer-Lambert law, we established an algorithm for a CW functional near-infrared spectroscopy (CW-fNIRS). The accuracy of this algorithm was validated both in comparison with direct measurements on brain tissue model and in vivo measurement upon human valsalva maneuver. This described method can also be utilized for other CW-fNIRS instruments to establish measuring algorithm.

Functional near-infrared spectroscopy (fNIRS) estimates the functional oscillations of oxyhemoglobin and deoxyhemoglobin in the cortex through scalp-located multiwavelength recordings. Hemoglobin oscillations are inferred through temporal changes in continuous-wave (CW) light attenuation. However, because of the diffusive multilayered head tissue structures, the photon path is longer than the source-detector separation, complicating hemoglobin evaluation. This aspect is incorporated in the modified Beer-Lambert law where the source-detector distance is multiplied by the differential pathlength factor (DPF). Since DPF estimation requires photons' time-of-flight information, DPF is assumed a priori in CW-fNIRS. Importantly, errors in the DPF spectrum induce hemoglobin cross talk, which is detrimental for fNIRS. We propose to estimate subject-specific DPF spectral dependence relying on multidistance high-density measurements. The procedure estimates the effective attenuation coefficient (EAC), which is proportional to the geometric mean of absorption and reduced scattering. Since DPF depends on the scattering-to-absorption ratio, EAC limits the spectral dependence assumption to scattering. This approach was compared to a standard frequency-domain multidistance procedure. A good association between the two methods ( ) was obtained. This approach could estimate low-resolution maps of the DPF spectral dependence through large field of view, high-density systems, reducing hemoglobin cross talk, and increasing fNIRS sensitivity and specificity to brain activity without instrumentation modification.

IntroductionOur present study set out to investigate the instant state of the prefrontal cortex (PFC) in healthy subjects before and after placement in the -12°head-down tilt (HDT) position in order to explore the mechanism behind the low-dissipation optimization state of the PFC.Methods40 young, right-handed healthy subjects (male: female = 20: 20) were enrolled in this study. Three resting state positions, 0°initial position, -12°HDT position, and 0°rest position were sequentially tested, each for 10 minutes. A continuous-wave functional near-infrared spectroscopy (fNIRS) instrument was used to assess the resting state hemodynamic data of the PFC. After preprocessing the hemodynamics data, we evaluated changes in resting-state functional connectivity (rsFC) level and beta values of PFC. The subjective visual analogue scale (VAS) was applied before and after the experiment. The presence of sleep changes or adverse reactions were also recorded.ResultsPairwise comparisons of the concentrations of oxyhemoglobin (HbO), deoxyhemoglobin (HbR), and hemoglobin (HbT) revealed significant differences in the aforementioned positions. Specifically, the average rsFC of PFC showed a gradual increase throughout the whole process. In addition, based on graph theory, the topological properties of brain network, such as small-world network and nodal degree centrality were analyzed. The results show that global efficiency and small-world sigma (σ) value were differences between 0°initial and 0°rest.DiscussionIn this study, placement in the -12°HDT had a significant effect on PFC function, mainly manifested as self-inhibition, decreased concentration of HbO in the PFC, and improved rsFC, which may provide ideas to the understanding and explanation of neurological diseases.

Since the frequency of the low-frequency oscillation signal [1] (LFO) of the blood oxygen is 0.01 Hz to 0.15 Hz and it is independent of the heart beat signal and the respiratory signal, the slow change of the hemodynamic parameters can be detected. In order to facilitate the low frequency oscillation signal to the non-hair-bearing forehead, we have developed a frequency-division-multiplexed modulated continuous wave functional near-infrared spectroscopy imaging system [2]. The light source emits laser light of the same wavelength at the same time, and the carrier frequency of the emitted light source is different at the same time [3]. This method is called frequency division multiplexing and can effectively eliminates crosstalk between light sources [4].

HighlightsWhat are the main findings?A seed-based resting-state functional connectivity (RSFC) and motor-paradigm-based oxygenation change assessment using continuous-wave functional near-infrared spectroscopy is feasible in patients with acute intracerebral hemorrhage (ICH).In patients with left hemispheric ICH, RSFC may be increased between the affected primary motor cortex (priMC) and the affected premotor cortex (preMC). In contrast, in right hemispheric ICH, RSFC may be decreased between the unaffected priMC and the affected somatosensory cortex.In patients with right hemispheric ICH with left hand finger tapping, there may be increased oxygenation over the unaffected preMC.What is the implication of the main finding?Motor cortex reorganization in patients with acute ICH is based on the side of the stroke.Left preMC connectivity and activity may be affected early in patients with ICH, which may serve as a target for neuromodulation devices.Insights into motor cortex remodeling may enable the development of more effective rehabilitation strategies during the acute phase. We aim to assess the affected and unaffected motor/premotor/somatosensory cortex resting state functional connectivity (RSFC) and reactivity with continuous wave functional near-infrared spectroscopy (cw-fNIRS) in patients with ICH compared to age, sex, and comorbidity-matched subjects. We enrolled patients with acute–subacute hemispheric ICH (n = 37; two were excluded due to artifacts) and grouped them according to the side (right and left) of the stroke. Matched participants or patients with recent transient ischemic attack were enrolled as control subjects for the study (n = 44; five were excluded due to artifacts). RSFC was assessed in both affected and unaffected hemispheres by group-level seed-based (primary motor cortex, priMC) correlation analysis. FT-associated relative oxyhemoglobin (ΔHbO) changes were analyzed in affected and unaffected hemispheres with generalized linear model regression. In left hemispheric ICH, the resting state coherence between the affected priMC and the affected premotor cortex (preMC) increased (β = 0.83, 95% CI = 0.19, 1.47, p = 0.01). In contrast, in right hemispheric ICH, the coherence between the unaffected priMC and the affected preMC decreased (β = −0.6, 95% CI = −1.12, −0.09, p = 0.02). In the left hemispheric ICH, the left-hand FT was associated with increased ΔHbO over the affected preMC (β = 0.01, 95% CI = 0.003, 0.02, p = 0.01). In contrast, in right hemispheric ICH, the left-hand FT was associated with increased ΔHbO over the unaffected preMC (β = 0.02, 95% CI = 0.006, 0.04, p = 0.01). Left hemispheric preMC may be involved in motor cortex reorganization in acute ICH in either hemisphere. Further studies may be required to assess longitudinal changes in motor cortex reorganization to inform acute motor rehabilitation.

The international standard IEC 80601-2-71 defines requirements for the basic safety and essential performance of medical continuous wave functional near-infrared spectroscopy (fNIRS) equipment, including a set of tests based on a dedicated tissue-equivalent phantom consisting of switchable apertures sandwiched between two diffusing media. In a companion paper, we described a first implementation of this phantom developed by AIST, its basic characterization of attenuation properties, and results of its application in the testing of commercial fNIRS devices. In the present paper, we provide the absorption and reduced scattering spectra of the AIST phantom and of alternative silicone phantom matrices, complemented with extensive simulations on the effect of the phantom optical properties and geometry on the features of the phantom relevant for the fNIRS standard test. All spectral measurements were performed with time-domain diffuse optical spectrometers for accurate quantification of the absorption and reduced scattering coefficient, whereas simulations were obtained using a Monte Carlo code. The POM-C matrix of the AIST phantom exhibits negligible absorption ( ) and a monotonically decreasing reduced scattering coefficient (from 10.5 down to ) in the 650 to 850nm range. Monte Carlo simulations explain the independence from wavelength, geometry, and partially also optical properties in the change in attenuation when switching between two apertures of 6 and 4mm diameter within the phantom, with spectral variation. Extended spectral characterization of POM-C and silicone matrices in the 600 to 1100nm range showed additional marked absorption peaks around 900 and 1020nm. We provide valid support for researchers and manufacturers to understand the operation of the phantom described in the fNIRS standard and to replicate it in their labs. In addition, we provide a basis for further evolution of the standard itself to encompass alternative probe geometries or functional tests.

.SignificanceWe present a fiberless, portable, and modular continuous wave-functional near-infrared spectroscopy system, Spotlight, consisting of multiple palm-sized modules—each containing high-density light-emitting diode and silicon photomultiplier detector arrays embedded in a flexible membrane that facilitates optode coupling to scalp curvature.AimSpotlight’s goal is to be a more portable, accessible, and powerful functional near-infrared spectroscopy (fNIRS) device for neuroscience and brain–computer interface (BCI) applications. We hope that the Spotlight designs we share here can spur more advances in fNIRS technology and better enable future non-invasive neuroscience and BCI research.ApproachWe report sensor characteristics in system validation on phantoms and motor cortical hemodynamic responses in a human finger-tapping experiment, where subjects wore custom 3D-printed caps with two sensor modules.ResultsThe task conditions can be decoded offline with a median accuracy of 69.6%, reaching 94.7% for the best subject, and at a comparable accuracy in real time for a subset of subjects. We quantified how well the custom caps fitted to each subject and observed that better fit leads to more observed task-dependent hemodynamic response and better decoding accuracy.ConclusionsThe advances presented here should serve to make fNIRS more accessible for BCI applications.

In this work, we investigate some major issues for the use of silicon photomultiplier (SiPM) devices in continuous wave functional near-infrared spectroscopy (CW fNIRS). We analyzed the after-pulsing effect, proposing the physical mechanism causing it, and determining its relevance for CW fNIRS. We studied the SiPM transients occurring as the SiPM device goes from the dark (LED in off state) to the illumination (LED in on state) conditions, and vice-versa. Finally, we studied the SiPM SNR in standard CW fNIRS operation.

PurposePsycholinguistic models traditionally view verbal working memory capacity as independent from linguistic features; connectionist models suggest otherwise. Moreover, lexical processing studies show high frequency words differ in cognitive effort from low frequency words, although these effects during concurrent processing of words in working memory are unknown. This novel study examines potential differences in cognitive effort, as measured by differences in HbO2 and Hb, for high frequency versus low frequency words during a working memory paradigm.MethodsA total of 21 neurologically typical participants (age 18–23) completed an auditory, n-back, working memory task comparing performance with high- as compared to low- frequency words. Hemodynamic changes in the prefrontal cortex were recorded with a continuous-wave functional near-infrared spectroscopy (fNIRS) device. Behavioral data (accuracy, reaction time) were recorded using E-prime.ResultsDifferences in word frequency were evident at both behavioral and neurological levels. Participants were more accurate, albeit slower in identifying the target two back in a sequence for low- as compared to high-frequency words. Patterns of hemodynamic changes were also significantly different between HF and LF conditions.ConclusionThe results from this study indicate that the behavioral and neurological signatures inherent in holding high- versus low-frequency words in working memory differs significantly. Specifically, the findings from this study indicated that words differing in frequency place different demands on cognitive processing load in memory updating tasks.

Background/Objectives: Lightweight and portable functional near-infrared spectroscopy (fNIRS) systems enable neuromonitoring in clinical environments such as operating rooms. Patient posture is known to influence physiology, behavior, and brain activity, and may affect fNIRS measurements. However, the effects of some postures commonly used in clinical care-such as Fowler's and semi-Fowler's-remain largely unexamined in fNIRS research. Methods: We conducted a singular study in a mock operating room exploring the effects of five postures-standing, upright sitting, Fowler's, semi-Fowler's, and supine-on fNIRS data during resting-state conditions and under various auditory stimuli. We collected hemodynamic data and extracted the characteristic hemodynamic response function (HRF) at each posture in response to the presented auditory stimulus and the amplitude of the resting-state signal. Results: For the auditory task condition, we found that posture had no statistically significant impact on the amplitude of the global HRF for Fowler's and semi-Fowler's postures. We also found no significant relationships across different postures when analyzing the amplitude of the global resting-state signal; however, binning of frequency-dependent postural effects revealed statistically significant differences between Fowler's and semi-Fowler's postures at low frequencies (f < 0.09 Hz). Conclusions: Our results suggest posture effects need not require complex data processing pipelines or data segmentation efforts on an auditory task-induced condition or on the general analysis of the global resting signal; however, not all reclined postures are equivalent, and we recommend that researchers report the angle of reclination measurements for seated data collection sessions for improved reliability and data context.

Continuous-wave near-infrared spectroscopy and imaging enable tissue hemodynamics and oxygenation to be determined non-invasively. Movements of the investigated subject can cause movement artifacts (MAs) in the recorded signals. The strength and type of MAs induced depend on the measurement principle. The aim of the present study was to investigate the quantitative relationship between different single-distance (SD) and multi-distance (MD) measurement methods and their susceptibility to MAs. We found that each method induces MAs to a different degree, and that MD methods are more robust against MAs than SD methods.

Objective. Memory is formed by the interaction of various brain functions at the item and task level. Revealing individual and combined effects of item- and task-related processes on retrieving episodic memory is an unsolved problem because of limitations in existing neuroimaging techniques. To investigate these issues, we analyze fast and slow optical signals measured from a custom-built continuous wave functional near-infrared spectroscopy (CW-fNIRS) system. Approach. In our work, we visually encode the words to the subjects and let them recall the words after a short rest. The hemodynamic responses evoked by the episodic memory are compared with those evoked by the semantic memory in retrieval blocks. In the fast optical signal, we compare the effects of old and new items (previously seen and not seen) to investigate the item-related process in episodic memory. The Kalman filter is simultaneously applied to slow and fast optical signals in different time windows. Main results. A significant task-related HbR decrease was observed in the episodic memory retrieval blocks. Mean amplitude and peak latency of a fast optical signal are dependent upon item types and reaction time, respectively. Moreover, task-related hemodynamic and item-related fast optical responses are correlated in the right prefrontal cortex. Significance. We demonstrate that episodic memory is retrieved from the right frontal area by a functional connectivity between the maintained mental state through retrieval and item-related transient activity. To the best of our knowledge, this demonstration of functional NIRS research is the first to examine the relationship between item- and task-related memory processes in the prefrontal area using single modality.

Continuous wave functional near-infrared spectroscopy: Various signal components and appropriate management

In the late 1980s and early 1990s, Dr. Britton Chance and his colleagues, using picosecond-long laser pulses, spearheaded the development of time-resolved spectroscopy techniques in an effort to obtain quantitative information about the optical characteristics of the tissue. These efforts by Chance and colleagues expedited the translation of near-infrared spectroscopy (NIRS)-based techniques into a neuroimaging modality for various cognitive studies. Beginning in the early 2000s, Dr. Britton Chance guided and steered the collaboration with the Optical Brain Imaging team at Drexel University toward the development and application of a field deployable continuous wave functional near-infrared spectroscopy (fNIR) system as a means to monitor cognitive functions, particularly during attention and working memory tasks as well as for complex tasks such as war games and air traffic control scenarios performed by healthy volunteers under operational conditions. Further, these collaborative efforts led to various clinical applications, including traumatic brain injury, depth of anesthesia monitoring, pediatric pain assessment, and brain–computer interface in neurology. In this paper, we introduce how these collaborative studies have made fNIR an excellent candidate for specified clinical and research applications, including repeated cortical neuroimaging, bedside or home monitoring, the elicitation of a positive effect, and protocols requiring ecological validity. This paper represents a token of our gratitude to Dr. Britton Chance for his influence and leadership. Through this manuscript we show our appreciation by contributing to his commemoration and through our work we will strive to advance the field of optical brain imaging and promote his legacy.