Source-detector Separation Research Articles

Coagulation depth estimation using a line scanner for depth-resolved laser speckle contrast imaging.

Partial-thickness burn wounds extend partially through the dermis, leaving many pain receptors intact and making the injuries very painful. Due to the painfulness, quick assessment of the burn depth is important to not delay surgery of the wound if needed. Laser speckle imaging (LSI) of skin blood flow can be helpful in finding severe coagulation zones with impaired blood flow. However, LSI measurements are typically too superficial to properly reach the full depth of the adult dermis and cannot resolve the flow in depth. Diffuse correlation spectroscopy (DCS) uses varying source-detector separations to allow differentiation of flow depths but requires time-consuming 2D scanning to form an image of the burn area. We here present a prototype for a hybrid DCS and LSI technique called speckle contrast diffuse correlation spectroscopy (scDCS) with the novel approach of using a laser line as a source and using the speckle contrast of averaged images to obtain an estimate of static scattering in the tissue. This will allow for fast non-contact 1D scanning to perform 3D tomographic imaging, making quantitative estimates of the depth and area of the coagulation zone from burn wounds. Simulations and experimental results from a volumetric flow phantom and a gelatin wedge phantom show promise to determine coagulation depth. The aim is to develop a method that, in the future, could provide more quantitative estimates of coagulation depth in partial thickness burn wounds to better estimate when surgery is needed.

Time-domain diffuse correlation spectroscopy at large source detector separation for cerebral blood flow recovery.

Time-domain diffuse correlation spectroscopy (td-DCS) enables the depth discrimination in tissue's blood flow recovery, considering the fraction of photons detected with higher time of flight (TOF) and longer pathlength through the tissue. However, the recovery result depends on factors such as the instrument response function (IRF), analyzed TOF gate start time, gate width and the source-detector separation (SDS). In this research we evaluate the performance of the td-DCS technique at three SDSs of 1.5, 2 and 2.5 cm to recover cerebral blood flow (CBF). To do that we presented comprehensive characterization of the td-DCS system through a series of phantom experiments. First by quality metrices such as coefficient of variation and contrast-to-noise ratios, we identified optimal time gate(s) of the TOF to extract dynamics of particles. Then using sensitivity metrices, each SDS ability to detect dynamics of particles in superficial and deeper layer was evaluated. Finally, td-DCS at each SDS was tested on healthy volunteers during cuff occlusion test and breathing tasks. According to phantom measurements, the sensitivity to estimate perfusion within the deep layer located at depth of 1.5 cm from the surface can be increased more than two times when the SDS increases from 1.5 cm to 2.5 cm.

Exploring the bias: how skin color influences oxygen saturation readings via Monte Carlo simulations.

Our goal is to understand the root cause of reported oxygen saturation ( ) overestimation in heavily pigmented skin types to devise solutions toward enabling equity in pulse oximeter designs. We aim to gain theoretical insights into the effect of skin tone on curves using a three-dimensional, four-layer tissue model representing a finger. A finger tissue model, comprising the epidermis, dermis, two arteries, and a bone, was developed using a Monte Carlo-based approach in the MCmatlab software. Two skin tones-light and dark-were simulated by adjusting the absorption and scattering properties within the epidermal layer. Following this, curves were generated in various tissue configurations, including transmission and reflection modes using red and infrared wavelengths. In addition, the influence of source-detector (SD) separation distances on both light and dark skin tissue models was studied. In transmission mode, curves did not deviate with changes in skin tones because both pulsatile and non-pulsatile terms experienced equal attenuation at red and infrared wavelengths. However, in reflection mode, measurable variations in curves were evident. This was due to differential attenuation of the red components, which resulted in a lower perfusion index at the red wavelength in darker skin. As the SD separation increased, the effect of skin tone on curves in reflection mode became less pronounced, with the largest SD separation exhibiting effects similar to those observed in transmission mode. Monte Carlo simulations have demonstrated that different light pathlengths within the tissue contribute to the overestimation of in people with darker skin in reflection mode pulse oximetry. Increasing the SD separation may mitigate the effect of skin tone on readings. These trends were not observed in transmission mode; however, further planned research using more complex models of the tissue is essential.

Application of the Single Source-Detector Separation Algorithm in Wearable Neuroimaging Devices: A Step toward Miniaturized Biosensor for Hypoxia Detection.

Most currently available wearable devices to noninvasively detect hypoxia use the spatially resolved spectroscopy (SRS) method to calculate cerebral tissue oxygen saturation (StO2). This study applies the single source-detector separation (SSDS) algorithm to calculate StO2. Near-infrared spectroscopy (NIRS) data were collected from 26 healthy adult volunteers during a breath-holding task using a wearable NIRS device, which included two source-detector separations (SDSs). These data were used to derive oxyhemoglobin (HbO) change and StO2. In the group analysis, both HbO change and StO2 exhibited significant change during a breath-holding task. Specifically, they initially decreased to minimums at around 10 s and then steadily increased to maximums, which were significantly greater than baseline levels, at 25-30 s (p-HbO < 0.001 and p-StO2 < 0.05). However, at an individual level, the SRS method failed to detect changes in cerebral StO2 in response to a short breath-holding task. Furthermore, the SSDS algorithm is more robust than the SRS method in quantifying change in cerebral StO2 in response to a breath-holding task. In conclusion, these findings have demonstrated the potential use of the SSDS algorithm in developing a miniaturized wearable biosensor to monitor cerebral StO2 and detect cerebral hypoxia.

Monte Carlo simulation of the effect of melanin concentration on light-tissue interactions in transmittance and reflectance finger photoplethysmography

Photoplethysmography (PPG) uses light to detect volumetric changes in blood, and is integrated into many healthcare devices to monitor various physiological measurements. However, an unresolved limitation of PPG is the effect of skin pigmentation on the signal and its impact on PPG based applications such as pulse oximetry. Hence, an in-silico model of the human finger was developed using the Monte Carlo (MC) technique to simulate light interactions with different melanin concentrations in a human finger, as it is the primary determinant of skin pigmentation. The AC/DC ratio in reflectance PPG mode was evaluated at source-detector separations of 1 mm and 3 mm as the convergence rate (Q), a parameter that quantifies the accuracy of the simulation, exceeded a threshold of 0.001. At a source-detector separation of 3 mm, the AC/DC ratio of light skin was 0.472 times more than moderate skin and 6.39 than dark skin at 660 nm, and 0.114 and 0.141 respectively at 940 nm. These findings are significant for the development of PPG-based sensors given the ongoing concerns regarding the impact of skin pigmentation on healthcare devices.

Skin temperature measurement based on diffuse reflection light at scattering variation independent source-detector separation

This study presents a method for predicting skin temperature using near-infrared light at 1314 nm and 1409 nm wavelengths. Skin temperature is a critical factor in the precision of in vivo optical measurements, necessitating accurate determination. Our approach detects temperature-induced tissue absorption changes, utilizing absorbance variations at specific scattering independent source-detector separations (SVI-SDSs) to estimate temperature. The 1409 nm wavelength is the main indicator for temperature prediction, with the 1314 nm wavelength used to identify the SVI-SDS for the 1409 nm light. Experimental data from four subjects were collected using an optical sensor across five SDSs ranging from 0.9 mm to 2.3 mm. We identified the SVI-SDS for the 1409 nm light at approximately 2.3 mm, where absorbance exhibited the most sensitivity to temperature changes, resulting in the lowest root mean square error (RMSE) and highest temperature sensitivity. These findings offer a valuable reference for the real-time monitoring and calibration of in vivo skin temperature, enhancing the accuracy of optical measurement techniques.

Application of transfer learning for rapid calibration of spatially resolved diffuse reflectance probes for extraction of tissue optical properties.

Treatment planning for light-based therapies including photodynamic therapy requires tissue optical property knowledge. This is recoverable with spatially resolved diffuse reflectance spectroscopy (DRS) but requires precise source-detector separation (SDS) determination and time-consuming simulations. An artificial neural network (ANN) to map from DRS at multiple SDS to optical properties was created. This trained ANN was adapted to fiber-optic probes with varying SDS using transfer learning (TL). An ANN mapping from measurements to Monte Carlo simulation to optical properties was created with one fiber-optic probe. A second probe with different SDS was used for TL algorithm creation. Data from a third were used to test this algorithm. The initial ANN recovered absorber concentration with (7.5% mean error) and at 665nm () with (2.5% mean error). For probe 2, TL significantly improved absorber concentration (0.38 versus RMSE, ) and (0.71 versus RMSE, ) recovery. A third probe also showed improved absorber (0.7 versus RMSE, ) and (1.68 versus RMSE, ) recovery. TL-based probe-to-probe calibration can rapidly adapt an ANN created for one probe to similar target probes, enabling accurate optical property recovery with the target probe.

Quantification of blood flow index in diffuse correlation spectroscopy using a robust deep learning method.

Diffuse correlation spectroscopy (DCS) is a powerful, noninvasive optical technique for measuring blood flow. Traditionally the blood flow index (BFi) is derived through nonlinear least-square fitting the measured intensity autocorrelation function (ACF). However, the fitting process is computationally intensive, susceptible to measurement noise, and easily influenced by optical properties (absorption coefficient and reduced scattering coefficient ) and scalp and skull thicknesses. We aim to develop a data-driven method that enables rapid and robust analysis of multiple-scattered light's temporal ACFs. Moreover, the proposed method can be applied to a range of source-detector distances instead of being limited to a specific source-detector distance. We present a deep learning architecture with one-dimensional convolution neural networks, called DCS neural network (DCS-NET), for BFi and coherent factor () estimation. This DCS-NET was performed using simulated DCS data based on a three-layer brain model. We quantified the impact from physiologically relevant optical property variations, layer thicknesses, realistic noise levels, and multiple source-detector distances (5, 10, 15, 20, 25, and 30mm) on BFi and estimations among DCS-NET, semi-infinite, and three-layer fitting models. DCS-NET shows a much faster analysis speed, around 17,000-fold and 32-fold faster than the traditional three-layer and semi-infinite models, respectively. It offers higher intrinsic sensitivity to deep tissues compared with fitting methods. DCS-NET shows excellent anti-noise features and is less sensitive to variations of and at a source-detector separation of 30mm. Also, we have demonstrated that relative BFi (rBFi) can be extracted by DCS-NET with a much lower error of 8.35%. By contrast, the semi-infinite and three-layer fitting models result in significant errors in rBFi of 43.76% and 19.66%, respectively. DCS-NET can robustly quantify blood flow measurements at considerable source-detector distances, corresponding to much deeper biological tissues. It has excellent potential for hardware implementation, promising continuous real-time blood flow measurements.

A refined analytical model for reconstruction problems in diffuse reflectance spectroscopy

A refined analytical model of spatially resolved diffuse reflectance with small source-detector separations (SDSs) for the in vivo skin studies is proposed. Compared to the conventional model developed by Farrell et al., it accounts for the limited acceptance angle of the detector fiber. The refined model is validated in the wide range of optical parameters by Monte Carlo simulations of skin diffuse reflectance at SDSs of units of mm. Cases of uniform dermis and two-layered epidermis-dermis structures are studied. Higher accuracy of the refined model compared to the conventional one is demonstrated in the separate, constraint-free reconstruction of absorption and reduced scattering spectra of uniform dermis from the Monte Carlo simulated data. In the case of epidermis-dermis geometry, the recovered values of reduced scattering in dermis are overestimated and the recovered values of absorption are underestimated for both analytical models. Presumably, in the presence of a thin mismatched topical layer, only the effective attenuation coefficient of the bottom layer can be accurately recovered using a diffusion theory-based analytical model while separate reconstruction of absorption and reduced scattering fails due to the inapplicability of the method of images. These findings require implementation of more sophisticated models of light transfer in inhomogeneous media in the recovery algorithms.

Comparison of diffuse correlation spectroscopy analytical models for measuring cerebral blood flow in adults.

Although multilayer analytical models have been proposed to enhance brain sensitivity of diffuse correlation spectroscopy (DCS) measurements of cerebral blood flow, the traditional homogeneous model remains dominant in clinical applications. Rigorous in vivo comparison of these analytical models is lacking. We compare the performance of different analytical models to estimate a cerebral blood flow index (CBFi) with DCS in adults. Resting-state data were obtained on a cohort of 20 adult patients with subarachnoid hemorrhage. Data at 1 and 2.5cm source-detector separations were analyzed with the homogenous, two-layer, and three-layer models to estimate scalp blood flow index and CBFi. The performance of each model was quantified via fitting convergence, fit stability, brain-to-scalp flow ratio (BSR), and correlation with transcranial Doppler ultrasound (TCD) measurements of cerebral blood flow velocity in the middle cerebral artery (MCA). The homogeneous model has the highest pass rate (100%), lowest coefficient of variation (CV) at rest (median [IQR] at 1Hz of 0.18 [0.13, 0.22]), and most significant correlation with MCA blood flow velocities (, ) compared with both the two- and three-layer models. The multilayer model pass rate was significantly correlated with extracerebral layer thicknesses. Discarding datasets with non-physiological BSRs increased the correlation between DCS measured CBFi and TCD measured MCA velocities for all models. We found that the homogeneous model has the highest pass rate, lowest CV at rest, and most significant correlation with MCA blood flow velocities. Results from the multilayer models should be taken with caution because they suffer from lower pass rates and higher coefficients of variation at rest and can converge to non-physiological values for CBFi. Future work is needed to validate these models in vivo, and novel approaches are merited to improve the performance of the multimodel models.

Absorption and reduced scattering coefficients in epidermis and dermis from a Swedish cohort study.

Knowledge of optical properties is important to accurately model light propagation in tissue, but in vivo reference data are sparse. The aim of our study was to present in vivo skin optical properties from a large Swedish cohort including 3809 subjects using a three-layered skin model and spatially resolved diffuse reflectance spectroscopy (Periflux PF6000 EPOS). Diffuse reflectance spectra (475 to 850nm) at 0.4 and 1.2mm source-detector separations were analyzed using an inverse Monte Carlo method. The model had one epidermis layer with variable thicknesses and melanin-related absorptions and two dermis layers with varying hemoglobin concentrations and equal oxygen saturations. The reduced scattering coefficient was equal across all layers. Median absorption coefficients () in the upper dermis ranged from 0.094 at 475nm to 0.0048 at 850nm and similarly in the lower dermis from 0.059 to 0.0035. The reduced scattering coefficient () ranged from 3.22 to 1.20, and the sampling depth (mm) ranged from 0.23 to 0.38 (0.4mm separation) and from 0.49 to 0.68 (1.2mm separation). There were differences in optical properties across sex, age groups, and BMI categories. Reference material for skin optical properties is presented.

Time domain diffuse Raman spectroscopy using single pixel detection.

Diffuse Raman spectroscopy (DIRS) extends the high chemical specificity of Raman scattering to in-depth investigation of thick biological tissues. We present here a novel approach for time-domain diffuse Raman spectroscopy (TD-DIRS) based on a single-pixel detector and a digital micromirror device (DMD) within an imaging spectrometer for wavelength encoding. This overcomes the intrinsic complexity and high cost of detection arrays with ps-resolving time capability. Unlike spatially offset Raman spectroscopy (SORS) or frequency offset Raman spectroscopy (FORS), TD-DIRS exploits the time-of-flight distribution of photons to probe the depth of the Raman signal at a single wavelength with a single source-detector separation. We validated the system using a bilayer tissue-bone mimicking phantom composed of a 1 cm thick slab of silicone overlaying a calcium carbonate specimen and demonstrated a high differentiation of the two Raman signals. We reconstructed the Raman spectra of the two layers, offering the potential for improved and quantitative material analysis. Using a bilayer phantom made of porcine muscle and calcium carbonate, we proved that our system can retrieve Raman peaks even in the presence of autofluorescence typical of biomedical tissues. Overall, our novel TD-DIRS setup proposes a cost-effective and high-performance approach for in-depth Raman spectroscopy in diffusive media.

Enhancing diffuse correlation spectroscopy pulsatile cerebral blood flow signal with near-infrared spectroscopy photoplethysmography.

Combining near-infrared spectroscopy (NIRS) and diffuse correlation spectroscopy (DCS) allows for quantifying cerebral blood volume, flow, and oxygenation changes continuously and non-invasively. As recently shown, the DCS pulsatile cerebral blood flow index () can be used to quantify critical closing pressure (CrCP) and cerebrovascular resistance (). Although current DCS technology allows for reliable monitoring of the slow hemodynamic changes, resolving pulsatile blood flow at large source-detector separations, which is needed to ensure cerebral sensitivity, is challenging because of its low signal-to-noise ratio (SNR). Cardiac-gated averaging of several arterial pulse cycles is required to obtain a meaningful waveform. Taking advantage of the high SNR of NIRS, we demonstrate a method that uses the NIRS photoplethysmography (NIRS-PPG) pulsatile signal to model DCS , reducing the coefficient of variation of the recovered pulsatile waveform () and allowing for an unprecedented temporal resolution (266Hz) at a large source-detector separation (). In 10 healthy subjects, we verified the quality of the NIRS-PPG during common tasks, showing high fidelity against ( ). We recovered CrCP and at 0.25Hz, times faster than previously achieved with DCS. NIRS-PPG improves DCS SNR, reducing the number of gate-averaged heartbeats required to recover CrCP and .

Evaluating feasibility of functional near-infrared spectroscopy in dolphins.

Using functional near-infrared spectroscopy (fNIRS) in bottlenose dolphins (Tursiops truncatus) could help to understand how echolocating animals perceive their environment and how they focus on specific auditory objects, such as fish, in noisy marine settings. To test the feasibility of near-infrared spectroscopy (NIRS) in medium-sized marine mammals, such as dolphins, we modeled the light propagation with computational tools to determine the wavelengths, optode locations, and separation distances that maximize sensitivity to brain tissue. Using frequency-domain NIRS, we measured the absorption and reduced scattering coefficient of dolphin sculp. We assigned muscle, bone, and brain optical properties from the literature and modeled light propagation in a spatially accurate and biologically relevant model of a dolphin head, using finite-element modeling. We assessed tissue sensitivities for a range of wavelengths (600 to 1700nm), source-detector distances (50 to 120mm), and animal sizes (juvenile model 25% smaller than adult). We found that the wavelengths most suitable for imaging the brain fell into two ranges: 700 to 900nm and 1100 to 1150nm. The optimal location for brain sensing positioned the center point between source and detector 30 to 50mm caudal of the blowhole and at an angle 45deg to 90deg lateral off the midsagittal plane. Brain tissue sensitivity comparable to human measurements appears achievable only for smaller animals, such as juvenile bottlenose dolphins or smaller species of cetaceans, such as porpoises, or with source-detector separations in adult dolphins. Brain measurements in juvenile or subadult dolphins, or smaller dolphin species, may be possible using specialized fNIRS devices that support optode separations of . We speculate that many measurement repetitions will be required to overcome hemodynamic signals originating predominantly from the muscle layer above the skull. NIRS measurements of muscle tissue are feasible today with source-detector separations of 50mm, or even less.

Shortwave infrared diffuse optical wearable probe for quantification of water and lipid content in emulsion phantoms using deep learning.

The shortwave infrared (SWIR, to 2000nm) holds promise for label-free measurements of water and lipid content in thick tissue, owed to the chromophore-specific absorption features and low scattering in this range. In vivo water and lipid estimations have potential applications including the monitoring of hydration, volume status, edema, body composition, weight loss, and cancer. To the best of our knowledge, there are currently no point-of-care or wearable devices available that exploit the SWIR wavelength range, limiting clinical and at-home translation of this technology. To design and fabricate a diffuse optical wearable SWIR probe for water and lipid quantification in tissue. Simulations were first performed to confirm the theoretical advantage of SWIR wavelengths over near infrared (NIR). The probe was then fabricated, consisting of light emitting diodes at three wavelengths (980, 1200, 1300nm) and four source-detector (S-D) separations (7, 10, 13, 16mm). In vitro validation was then performed on emulsion phantoms containing varying concentrations of water, lipid, and deuterium oxide (). A deep neural network was developed as the inverse model for quantity estimation. Simulations indicated that SWIR wavelengths could reduce theoretical water and lipid extraction errors from to when compared to NIR wavelengths. The SWIR probe had good signal-to-noise ratio ( up to 10mm S-D) and low drift ( up to 10mm S-D). Quantification error in emulsion phantoms was for water and for lipid. Water estimation during a dilution experiment had an error of . This diffuse optical SWIR probe was able to quantify water and lipid contents in vitro with good accuracy, opening the door to human investigations.

Portable, high speed blood flow measurements enabled by long wavelength, interferometric diffuse correlation spectroscopy (LW-iDCS)

Diffuse correlation spectroscopy (DCS) is an optical technique that can be used to characterize blood flow in tissue. The measurement of cerebral hemodynamics has arisen as a promising use case for DCS, though traditional implementations of DCS exhibit suboptimal signal-to-noise ratio (SNR) and cerebral sensitivity to make robust measurements of cerebral blood flow in adults. In this work, we present long wavelength, interferometric DCS (LW-iDCS), which combines the use of a longer illumination wavelength (1064 nm), multi-speckle, and interferometric detection, to improve both cerebral sensitivity and SNR. Through direct comparison with long wavelength DCS based on superconducting nanowire single photon detectors, we demonstrate an approximate 5× improvement in SNR over a single channel of LW-DCS in the measured blood flow signals in human subjects. We show equivalence of extracted blood flow between LW-DCS and LW-iDCS, and demonstrate the feasibility of LW-iDCS measured at 100 Hz at a source-detector separation of 3.5 cm. This improvement in performance has the potential to enable robust measurement of cerebral hemodynamics and unlock novel use cases for diffuse correlation spectroscopy.

Implementation of data fusion to increase the efficiency of classification of precancerous skin states using in vivo bimodal spectroscopic technique.

This study presents the results of the classification of diffuse reflectance (DR) spectra and multiexcitation autofluorescence (AF) spectra that were collected in vivo from precancerous and benign skin lesions at three different source detector separation (SDS) values. Spectra processing pipeline consisted of dimensionality reduction, which was performed using principal component analysis (PCA), followed by classification step using such methods as support vector machine (SVM), multilayered perceptron (MLP), linear discriminant analysis (LDA), and random forest (RF). In order to increase the efficiency of lesion classification, several data fusion methods were applied to the classification results: majority voting, stacking, and manual optimization of weights. The results of the study showed that in most of cases the use of data fusion methods increased the average multiclass classification accuracy from 2% up to 4%. The highest accuracy of multiclass classification was obtained using the manual optimization of weights and reached 94.41%.

Novel data types for frequency-domain diffuse optical spectroscopy and imaging of tissues: characterization of sensitivity and contrast-to-noise ratio for absorption perturbations.

In frequency-domain (FD) diffuse optics it is known that the phase of photon-density waves (ϕ) has a stronger deep-to-superficial sensitivity ratio to absorption perturbations than the alternate current (AC) amplitude, or the direct current intensity (DC). This work is an attempt to find FD data types that feature similar or even better sensitivity and/or contrast-to-noise for deeper absorption perturbations than phase. One way is to start from the definition of characteristic function (Xt(ω)) of the photon's arrival time (t) and combining the real () and imaginary parts () with phase to yield new data types. These new data types enhance the role of higher order moments of the probability distribution of the photon's arrival time t. We study the contrast-to-noise and sensitivity features of these new data types not only in the single-distance arrangement (traditionally used in diffuse optics), but we also consider the spatial gradients, which we named dual-slope arrangements. We have identified six data types that for typical values of the optical properties of tissues and depths of interest, have better sensitivity or contrast-to-noise features than phase data and that can be used to enhance the limits of imaging of tissue in FD near infrared spectroscopy (NIRS). For example, one promising data type is which shows, in the single-distance source-detector arrangement, an increase of deep-to-superficial sensitivity ratio with respect to phase by 41% and 27% at a source-detector separation of 25 and 35 mm, respectively. The same data type also shows an increase of contrast-to noise up to 35% with respect to phase when the spatial gradients of the data are considered.

Interstitial null-distance time-domain diffuse optical spectroscopy using a superconducting nanowire detector.

Interstitial fiber-based spectroscopy is gaining interest for real-time in vivo optical biopsies, endoscopic interventions, and local monitoring of therapy. Different from other photonics approaches, time-domain diffuse optical spectroscopy (TD-DOS) can probe the tissue at a few cm distance from the fiber tip and disentangle absorption from the scattering properties. Nevertheless, the signal detected at a short distance from the source is strongly dominated by the photons arriving early at the detector, thus hampering the possibility of resolving late photons, which are rich in information about depth and absorption. To fully benefit from the null-distance approach, a detector with an extremely high dynamic range is required to effectively collect the late photons; the goal of our paper is to test its feasibility to perform TD-DOS measurements at null source-detector separations (NSDS). In particular, we demonstrate the use of a superconducting nanowire single photon detector (SNSPD) to perform TD-DOS at almost NSDS by exploiting the high dynamic range and temporal resolution of the SNSPD to extract late arriving, deep-traveling photons from the burst of early photons. This approach was demonstrated both on Monte Carlo simulations and on phantom measurements, achieving an accuracy in the retrieval of the water spectrum of better than 15%, spanning almost two decades of absorption change in the 700- to 1100-nm range. Additionally, we show that, for interstitial measurements at null source-detector distance, the scattering coefficient has a negligible effect on late photons, easing the retrieval of the absorption coefficient. Utilizing the SNSPD, broadband TD-DOS measurements were performed to successfully retrieve the absorption spectra of the liquid phantoms. Although the SNSPD has certain drawbacks for use in a clinical system, it is an emerging field with research progressing rapidly, and this makes the SNSPD a viable option and a good solution for future research in needle guided time-domain interstitial fiber spectroscopy.

Neural network-based optimization of sub-diffuse reflectance spectroscopy for improved parameter prediction and efficient data collection.

In this study, a general and systematical investigation of sub-diffuse reflectance spectroscopy is implemented. A Gegenbauer-kernel phase function-based Monte Carlo is adopted to describe photon transport more efficiently. To improve the computational efficiency and accuracy, two neural network algorithms, namely, back propagation neural network and radial basis function neural network are utilized to predict the absorption coefficient , reduced scattering coefficient and sub-diffusive quantifier , simultaneously, at multiple source-detector separations (SDS). The predicted results show that the three parameters can be predicated accurately by selecting five SDSs or above. Based on the simulation results, a four wavelength (520, 650, 785 and 830 nm) measurement system using five SDSs is designed by adopting phase-lock-in technique. Furtherly, the trained neural-network models are utilized to extract optical properties from the phantom and in vivo experimental data. The results verify the feasibility and effectiveness of our proposed system and methods in mucosal disease diagnosis.