Neurophotonics最新文献

Significance: Optical functional neuroimaging relies upon accurate anatomical models to provide optimal data registration and image reconstruction.

Aim: We establish and validate a robust photogrammetric algorithm for scalp morphology that utilizes 3-dimensional (3D) imaging with a photogrammetric cap to provide individualized scalp morphology estimation through hair in the absence of magnetic resonance imaging (MRI).

Approach: The scalp morphology estimation uses a flexible, neoprene cap and 3D-printed photogrammetric modules with fiducials. We create a sparse scalp sampling and align the MNI152 atlas to generate the scalp morphology estimation. We used the international 10 to 20 electroencephalogram positions for alignment and calculated the error as the Euclidean distance among a subspace of modified 10 to 5 electroencephalogram points between surface-based methods and participant-specific MRI.

Results: The scalp morphology estimation error relative to participant-specific MRI had a mean (std) error of 4.27 (2.15) mm, a gold standard volumetric registration of 3.57 (1.69) mm, a four-point scaled atlas of 11.45 (6.00) mm, an unscaled atlas of 5.35 (1.52) mm, and a scalp estimation without cap of 12.42 (6.45) mm. Notably, the scalp morphology estimation demonstrated lower variance in the spatial distribution of error, indicating robustness to idiosyncratic head shapes.

Conclusions: Our scalp morphology estimation algorithm is robust in the presence of hair, provides accurate participant-specific head shapes without requiring MRI, and scales to other cap and fiducial designs, thus highlighting the utility of this tool for a variety of applications.

Significance: Functional near-infrared spectroscopy (fNIRS) enables neuroimaging in scenarios where other modalities are less suitable, such as during motion tasks or in low-resource environments. Sparse fNIRS arrays with 30 mm channel spacing are widely used but have limited spatial resolution. High-density (HD) arrays with overlapping, multidistance channels improve sensitivity and localization but increase costs and setup times. A statistical comparison of HD and sparse arrays is needed for evaluating the benefits and trade-offs of HD arrays.

Aim: This study provides a statistical comparison of HD and sparse fNIRS performance to inform array selection in future research.

Approach: We measured prefrontal cortex (PFC) activation during congruent and incongruent word-color Stroop (WCS) tasks using both sparse and HD arrays for 17 healthy adult participants, comparing dorsolateral PFC channel and image results at the group level.

Results: Although both arrays detected activation in channel space during incongruent WCS, channel and image space results demonstrated superior localization and sensitivity with the HD array for all WCS.

Conclusion: Sparse channel data may suitably detect activation from cognitively demanding tasks, such as incongruent WCS. However, the HD array outperformed sparse in detecting and localizing brain activity in image space, particularly during lower cognitive load tasks, making it more suitable for neuroimaging applications.

Significance: Early language acquisition represents a fundamental achievement in cognitive development, yet the neural mechanisms underlying this process remain debated, particularly whether specialized language regions exist from early life or emerge gradually through development.

Aim: We aim to investigate the functional specialization for language processing in early childhood. We first aimed to validate an individual functional channel of interest (fCOI) approach for dissociating language and cognitive control regions in adults and then to apply this method to examine whether these functional profiles are present in toddlers.

Approach: Using functional near-infrared spectroscopy with the fCOI approach, we conducted two experiments involving adults ( $N=20$ , ages 18 to 26 years) and toddlers ( $N=22$ , ages 2 to 4 years) who completed language processing (intact versus degraded speech) and cognitive control tasks (spatial working memory task for adults, go/no-go task for toddlers).

Results: For language regions within the left inferior frontal gyrus (LIFG), both adults and toddlers showed a significantly stronger response to intact versus degraded speech, with no significant modulation by cognitive demand manipulation. However, language selectivity in the homologous right hemisphere region was present only in adults. The multiple demand regions showed complementary patterns, with selectivity for cognitive control of regions within the right inferior frontal gyrus (RIFG) emerging early.

Conclusions: These findings provide evidence for early neural specialization of language processing in LIFG while revealing ongoing development in RIFG organization. Our results support models of early language-specific neural regions rather than gradual differentiation from domain-general mechanisms while highlighting the protracted development of language organization.

Significance: Real-time monitoring of neurovascular coupling (NVC) is crucial for early diagnosis and effective treatment strategies in neonates with hypoxic ischemic encephalopathy (HIE). In our previous studies, the NVC of neonates with HIE was determined using wavelet transform coherence (WTC) between the amplitude-integrated electroencephalogram (aEEG) and regional cerebral oxygen saturation ( ${\mathrm{SctO}}_2$ ) using a post-acquisition analysis.

Aim: We propose a time-resolved WTC analysis, providing a real-time analysis tool that facilitates immediate and continuous evaluation of cerebral hemodynamics and neuronal activity.

Approach: The real-time WTC framework employs a progressive zero-padding strategy with incremental temporal data integration. Initial analysis preserves 4 h of $\mathrm{aEEG}/{\mathrm{SctO}}_2$ data while zero-padding 16 h to maintain a 20-h window. This enables calculation of time-resolved significant coherence (trSC) at time 2 h (1- to 2-h window) within the 20- to 150-min scale range. The system subsequently advances hourly, preserving an additional hour of acquired data while proportionally reducing zero-padding. This cascading approach continues until full 20-h data preservation, with final trSC calculations targeting time 18 h (17- to 18-h window).

Results: We included 55 neonates with mild to severe HIE, the time-scale maps of which were obtained using both post-acquisition and real-time WTC analysis methods. Accordingly, trSC curves within the 20- to 150-min wavelet scale were statistically compared between the two methods using a linear mixed-effects model. There was no significant difference in trSC results between the two methods ( $p=0.159$ ). In addition, NVC was significantly lower in the moderate to severe HIE group compared with the mild HIE group at hours 3 and 4 ( $p<0.01$ ).

Conclusions: We demonstrated the feasibility of real-time dynamic WTC analysis for dynamic NVC in newborns with HIE, providing a potential bedside tool for the early detection of brain abnormalities.

Significance: Current preclinical evaluation of Alzheimer's disease pathology in mouse models relies on post-mortem analyses, which hinders the development and optimization of therapeutic approaches. Although in vivo methods exist, monitoring amyloid plaque signals across multiple brain regions in freely behaving animals remains a significant challenge.

Aim: We aim to develop an optical approach to address this challenge.

Approach: We used flat and tapered optical fibers in an Alzheimer's mouse model.

Results: We first confirmed that conventional flat fiber-based photometry can detect amyloid plaque signals across multiple brain regions under anesthesia after injecting a blood-brain-barrier-permeable tracer, Methoxy-X04. The depth profile of in vivo fluorescent signals is correlated with histological signals. A machine learning approach could distinguish between in vivo fluorescent signals of mice with and without amyloid plaques. Next, after validating the feasibility of depth-resolved fiber photometry ex vivo, we chronically implanted a tapered fiber to monitor amyloid plaque signals in freely behaving mice. After injecting Methoxy-X04, fluorescent signals increased in a depth-specific manner in Alzheimer's mice, but not in their wild-type littermates.

Conclusions: Our approach expands the capabilities of fiber photometry to monitor molecular pathologies, such as amyloid plaques, even in a freely behaving condition.

Professor Chris Xu reflects on his journey from a curious student in China to a leading researcher at Cornell University, offering insights into the evolution of his work, the mentors who shaped him, and the future of brain imaging.

Significance: Deep brain stimulation (DBS) is an established treatment for movement disorders and other neurological conditions. Accurate localization of small deep brain nuclei, such as the subthalamic nucleus (STN) and internal pallidum (GPi), is crucial for successful DBS outcomes. However, magnetic resonance imaging (MRI), commonly used for DBS planning, lacks the resolution and contrast needed to directly delineate these target structures.

Aim: We aim to explore the potential of catheter-based polarization-sensitive optical coherence tomography (PS-OCT) as a complementary imaging tool for high-resolution visualization of tissue surrounding the DBS insertion trajectory.

Approach: We simulated DBS implantation surgery at three targets in a post-mortem nonhuman primate head. PS-OCT, using advanced reconstruction algorithms for absolute depth-resolved birefringence, was compared with MRI for its ability to visualize and differentiate structural details.

Results: PS-OCT provided more detailed and accurate structural information than MRI while maintaining consistency with MRI results. Its compact form factor and imaging paradigm integrate seamlessly into the surgical workflow, offering new insights for intraoperative decision-making.

Conclusions: PS-OCT functions as an intraoperative imaging tool, offering valuable guidance during the procedure. These findings establish PS-OCT as a promising complementary tool for DBS, with potential for further clinical validation and in vivo studies.

Significance: Postnatal maternal anxiety affects a substantial number of new mothers and is linked to long-term risk for anxiety in their offspring. Yet, the neural mechanisms through which postnatal maternal anxiety influences early cognitive development remain unclear. We investigated whether postnatal maternal anxiety shapes how infant brains respond to unexpected events-prediction errors-which are central to learning in uncertain environments.

Aim: We examined prediction error processing in 6- to 8-month-old infants using high-density diffuse optical tomography and eye-tracking. We hypothesized that neural responses in the medial prefrontal cortex (mPFC) would vary with maternal anxiety levels.

Approach: Infants viewed audiovisual events where expected outcomes were occasionally omitted, eliciting prediction errors. Hemodynamic responses in the frontal cortex were analyzed using a general linear model, with trial-by-trial gaze data as a parametric modulator. Maternal anxiety was measured using the state-trait anxiety inventory.

Results: Prediction error responses were localized to the mPFC and were only detectable when controlling for infant attention using eye-tracking. Cortical activation in response to unexpected stimuli was significantly enhanced in infants of mothers with higher trait anxiety.

Conclusion: Our findings suggest that maternal anxiety modulates prediction error processing in the infant brain, potentially shaping early sensitivity to environmental unpredictability and conferring risk for later anxiety.

Significance: Multilayer (two- and three-layer) diffuse correlation spectroscopy (DCS) models improve cerebral blood flow index (CBFi) measurement sensitivity and mitigate interference from extracerebral tissues. However, their reliance on multiple predefined parameters (e.g., layer thickness and optical properties) and high computational load limit their feasibility for real-time bedside monitoring.

Aim: We aim to develop a fast, accurate DCS data processing method based on the two-layer DCS analytical model, enabling real-time cerebral perfusion monitoring with enhanced brain sensitivity.

Approach: We employed deep learning (DL) to accelerate DCS data processing. Unlike previous DCS networks trained on single-layer models, our network learns from the two-layer DCS analytical model, capturing extracerebral versus cerebral dynamics. Realistic noise was estimated from subject-specific baseline measurements using a $512\times 512$ SPAD array at a large source-detector separation (35 mm). The model was evaluated on test datasets simulated with a four-layer slab head model via Monte Carlo (MC) methods and compared against conventional single-exponential fitting and the two-layer analytical fitting. Two in vivo physiological response tests were also conducted to assess the real-world performance.

Results: The proposed method bypasses traditional curve-fitting and achieves real-time monitoring of CBF changes at 35 mm separation for the first time with a DL approach. Validation on MC simulations shows superior accuracy in relative CBFi estimation (4.1% error versus 12.7% for single-exponential fitting) and significantly enhanced CBFi sensitivity (86.5% versus 57.7%). Although the two-layer analytical fitting offers optimal performance, it depends on strict assumptions and preconditions, and its computational complexity makes it time-consuming and unsuitable for real-time monitoring applications.In addition, our method minimizes the influence of superficial blood flow and is 750-fold faster than single-exponential fitting in a realistic scenario. In vivo tests further validated the method's ability to support real-time cerebral perfusion monitoring and pulsatile waveform recovery.

Conclusions: This study demonstrates that integrating DL with the two-layer DCS analytical model enables accurate, real-time cerebral perfusion monitoring without sacrificing depth sensitivity. The proposed method enhances CBFi sensitivity and recovery accuracy, supporting future deployment in bedside neuro-monitoring applications.

Significance: Noninvasive optical measurements of blood flow have many applications. Measurements have been demonstrated with diffuse correlation spectroscopy (DCS), interferometric diffusing wave spectroscopy (iDWS), and speckle contrast optical spectroscopy (SCOS) techniques, but concurrent measurements with all three techniques in the same experiment have not been compared.

Aim: We aim to evaluate the comparative strengths and weaknesses of SCOS, iDWS, and DCS methods in controlled experiments.

Approach: We performed in vitro temperature-controlled microsphere flow phantom and in vivo arm cuff occlusion experiments using SCOS, iDWS, and DCS concurrently and in the same geometry.

Results: In vitro results showed absolute flow metrics agreement between iDWS and DCS and demonstrated large gains in signal-to-noise for iDWS and SCOS compared with DCS; relative changes in flow measured by SCOS were also in good agreement with DCS and iDWS. The in vivo cuff occlusion results showed agreement of relative changes in flow measured by DCS, iDWS, and SCOS. However, DCS recovered a flow pulsatility index that was larger than iDWS and SCOS indices.

Conclusions: The experiments demonstrate the equivalency of absolute flow measures from iDWS and DCS and improved precision of pulsatile waveforms from SCOS. These results emphasize the need for rapid development and adoption of iDWS and SCOS.