This PDF file contains the front matter for SPIE Proceedings Volume 12376, including the Title Page, Copyright information, Table of Contents and Conference Committee list.

Many studies on CW spatially resolved spectroscopy (CW-SRS) have been conducted to noninvasively determine the optical properties, particularly the absorption and reduced scattering coefficients, μa and μs′, of biological tissues. To determine both μa and μs′, conventional CW-SRS employs measurements of the diffuse reflectances at short source-detector (SD) distances in the non-diffusion regime. In contrast, CW-SRS with long SD distances in the diffusion regime can determine only the effective attenuation coefficients, μeff = (3μaμs′)1/2 without separating μa and μs′. This study proposes a new method to separately determine μa and μs′ using CW-SRS with long SD distances, extending to conditions with high and low internal reflection at the boundary of homogeneous semi-infinite media. The proposed method used two ratios of the diffuse reflectances at two long SD distances, and μa and μs′ were determined by fitting the theoretical ratios to the measured values. Numerical simulations were conducted to validate the proposed method. As a light propagation model, the analytical solution of the time-dependent photon diffusion equation under the partial-current boundary condition (TD-DE-PCBC), which is verified for high internal reflection, was employed. Simulated measurements of the two ratios were compared with the calculated ratios (so-called look-up tables) using the TD-DE-PCBC to determine both μa and μs of the media. Simulation results demonstrate the validity of the proposed method. The effects of deviations in the SD distances and internal reflection coefficients were evaluated. Changes in the light propagation paths in the medium are discussed, and methods to realize the proposed method are suggested.

In this paper, the accuracy of perturbation Monte Carlo(pMC) estimates as a function of scattering perturbation size for spatially-resolved diffuse reflectance over a broad range of optical properties is analyzed. We also propose a methodology to predict the variation of pMC variance with perturbation size based on the data from the reference simulation alone. The results show better pMC performance when based on results from reference Monte Carlo simulations that utilize a Russian Roulette as a variance reduction method. Specifically, we demonstrate that for a proximal detector we can estimate the pMC relative error within 5% of the true value for scattering perturbations in the range of [-15%, +20%]. For a distal, our method provides relative error estimates within 20% for scattering perturbations in the range of [-8%, +15%]. Moreover, improved performance is observed in case of both proximal and distal detectors when reference simulations performed at lower ( 𝜇𝑠 ′ 𝜇𝑎 ) values. This methodology is useful for the optimal design of pMC analysis of multi-spectral data sets.

We present the results of a validation study on the uniqueness of a new method of multiparameter spectrophotometry (MPS) without integrating spheres to determine radiation transfer (RT) parameters by measurement of 20% Intralipid samples. The new MPS method is based on a robust stochastic optimization algorithm combined with Monte Carlo simulation to model light matter interaction. Our results prove the uniqueness of the inverse solutions for the of MPS method, which can be further developed in easy-to-use instrument for determination of RT parameters of turbid samples.

Tissue mimicking phantoms are widely used to test and validate near-infrared spectroscopy (NIRS) devices and algorithms. The two main constituents of phantoms for NIRS applications are a light scatterer, typically Intralipid, and one or multiple light-absorbing dyes, which are most commonly methylene blue, indocyanine green (ICG), and India ink. The current study investigated the spectral shape of tissue mimicking phantoms made of different combinations of Intralipid and these three dyes. The results reveal that Intralipid interacts with the dyes and alters their molar extinction coefficients, thereby hindering the ability to accurately estimate the phantom chromophore concentrations when either the dye or scatterer concentrations change. Furthermore, inorganic light scatterers, such as glass microspheres and titanium oxide, have less interaction with the dyes, with glass microspheres being the least interactive and therefore the best scatterer. These findings are significant, because NIRS phantoms continue to depend on methylene blue, ICG, and India ink as chromophores. Therefore, glass microspheres should be used as a light scatterer instead of Intralipid, which is currently the most common scatterer for NIRS liquid phantoms.

Frequency domain (FD) diffuse optical spectroscopy (DOS) can be used to recover absolute optical properties of biological tissue, providing valuable clinical feedback, including in diagnosis and monitoring of breast tumours. In this study, tomographic (3D) and topographic (2D) techniques for spatially-varying optical parameter recovery are presented, based on a multi-distance, handheld DOS probe. Processing pipelines and reconstruction quality are discussed and quantitatively compared, demonstrating the trade-offs between depth sensitivity, optical contrast, and computational speed. Together, the two techniques provide both depth sensitive real-time feedback, and high-resolution 3D reconstruction from a single set of measurements, enabling faster and more accurate clinical feedback.

Frequency-domain near-infrared spectroscopy (FD-NIRS) can measure absolute tissue optical properties for functional brain imaging. Multiple source-detector separations (SDS), and multiple modulation frequencies can be used for FD-NIRS application in homogeneous or multi-layer tissue models. A multiple-frequency approach is advantageous in building more compact systems with a single SDS. In this work, we compare the accuracy of estimating absolute optical properties of a multi-layered tissue model using multi-distance and multi-frequency approaches. We demonstrate that the multiple-frequency approach is comparable in accuracy to the multiple-distance approach and can be confidently implemented for brain imaging applications.

Measurements of absolute optical properties of scattering samples is valuable in the field of bio-medicine and beyond. However, achieving these measurements is complicated by the need for calibration and by the large sample volumes typically needed to meet common diffusion theory models. We propose a method for calibration free absolute measurements of the absorption coefficient (µa) and reduced scattering coefficient (µ 0 s ) in a relatively small volume (the size of a standard cuvette; 45 mm × 10 mm × 10 mm). This method utilizes the previously proposed Self-Calibrating (SC) / Dual-Slope (DS) geometry by placing two light sources on one face of the cuvette and two optical detectors on the opposite face. This leads to the proposal of the Dual-Ratio of the complex Transmittance (Te) (DR{Te}), a method with the same advantages as SC / DS but with less geometric requirements and constraints. Here we confirm that measurement of DR{Te} from a cuvette may be converted to absolute optical properties. We then investigate differences between two choices of forward models for optical measurements in a cuvette, either Monte-Carlo or diffusion theory. A discrepancy between the two was found, which leads to an error of 10 % in µa and no error in µ'_s when Monte-Carlo was used to generate data and diffusion theory used to invert it. This result highlights the importance of which model is chosen for the inverse problem when this method is implemented in practice. Implementations would need to evaluate different models against ground truths to identify the optimal method for the measurement of absolute optical properties in a cuvette.

Frequency Domain (FD)-fNIRS workflow has been designed to demonstrate (i) benefits of phase in FD measurements to offers greater hemodynamic response function (HRF) contrast than intensity alone, (ii) multiple parameter recovery paradigms in dual slopes (DS) which use” nut”, not” banana” shaped sensitivity profiles and (iii) subject-atlas registration for source-detector locations and model-based tomography.

Frequency-domain (FD) optical tomography instruments modulate the intensity of the light source at a radio frequency and measure the amplitude and phase shift of the detected photon density wave. The differing spatial sensitivities of amplitude and phase to the optical properties of tissue suggest that inclusion of phase data can improve the image reconstruction accuracy. This study describes our methodology for improved use of FD data in conjunction with a Monte Carlo (MC) forward solver (Monte Carlo eXtreme; MCX) and a voxel-based model of a two-year-old child’s head. The child participated our previous study where subjects were stimulated with affective (slow brushing) and non-affective touch (fast brushing) to their right forearm, and the responses were measured from the left hemisphere with our in-house 16-channel high-density FD system. We implemented the computation of the FD sensitivity profiles to the MCX photon simulation software, and validated the output against our in-house MC code. We used simulated and the real experimental touch response data to observe the effects of including both FD data types to the image reconstruction instead of amplitude data alone. For the simulated and experimental case, we observed that the inclusion of phase data increases the reconstructed contrast in the brain. The individual touch responses showed similarity to the group-level results in our original publication with 16 subjects and amplitude data alone, and other literature.

Early treatment improves rheumatoid arthritis (RA) prognosis; however, 30% of patients still fail their first treatment and current methods can take as long as 3–6 months to detect treatment failure. Time-domain near-infrared diffuse optical imaging (TD-DOI) has the potential to track gradual changes in RA disease activity and could be a more sensitive technique for RA treatment monitoring. Nevertheless, there has been very little investigation into the relationships between TD-DOI and the specific physiological changes that occur in RA. To this end, this work’s objective was to investigate the effects of RA-associated physiological changes on TD-DOI images in silico. Virtual finger phantoms-derived from an MRI segmentation of a healthy human finger-were used to simulate changes in 5 physiological parameters that are typically affected in RA. Four levels of disease activity were considered for each parameter and each parameter was altered individually; parameter alterations were then translated into changes in phantom geometry and optical properties. MCXLAB was used to simulate the propagation of time-domain light (800 nm) through phantoms’ proximal interphalangeal joints and Poisson noise was added to the TD-DOI data to simulate experimental conditions. Spatiotemporal Fourier decomposition was applied to the TD-DOI data to extract image components, which were grouped into datasets based on the physiological parameter that was altered. Component sensitivity to each physiological parameter was assessed using the ratio between a component’s range and standard deviation within each dataset (range-to-error ratio; RER). Mean RER was highest for changes in synovial membrane and fluid volume, suggesting that this parameter may be a primary source of contrast for monitoring RA treatment response with TD-DOI.

In this article we introduce a systematic approach for optimal scanning of dynamically evolving objects, including cases where the dynamics is unknown. The method is specifically designed to optimize each measurement and engineer illumination patterns with the goal of reducing the uncertainty left in our estimation of the sample. Concurrently, the algorithm uses system identification techniques to develop a mathematical model for the dynamics under test based on the acquired data and it uses the model to predict changes in the distribution and optimize upcoming measurements. The theory is developed and simulations are provided to better display discussed concepts.

We demonstrate a motion-resistant, three-wavelength, spatial frequency domain imaging (SFDI) system with ambient light suppression using a new 8-tap CMOS image sensor developed in our laboratory. Compared to the previous sensor (134×150), the new sensor’s readout maximum frame rate has improved to 33fps from 6.28fps, and the new 700×540- pixel sensor allows imaging at a higher spatial resolution over a larger field of view. Furthermore, the number of projected images needed per wavelength is reduced from three to two after applying the Hilbert transform. One image of planar illumination and one image for sinusoidal pattern projection at three wavelengths as well as one image of ambient light are captured by the 8-tap image sensor concurrently. The bias caused by ambient light is removed by subtracting the ambient light image from other images. Suppression of motion artifacts is achieved by reducing the exposure and projection time of each pattern. Sufficient signal level is maintained by repeating the exposure multiple times. In this study, LEDs with wavelengths of 554nm, 660nm, and 730nm were used to estimate oxy-/deoxyhemoglobin and melanin concentrations from in-vivo volar forearm skin.

Time-resolved (TR) near-infrared spectroscopy (NIRS) is a promising technique for neuromonitoring, but there are currently very few TR-NIRS devices with the spectral range and resolution needed for accurate monitoring of cerebral blood oxygenation (StO2) and metabolism (cytochrome-c-oxidase; oxCCO). Here we present a hyperspectral TR compressive sensing spectrometer with a wide spectral range, high spectral resolution, and no after pulsing. A homogeneous blood-yeast phantom experiment was performed to evaluate the spectrometer’s ability to monitor S_tO₂ and oxCCO with and without compression. The effect of using a 90% compression rate on the recovered changes in deoxyhemoglobin (Hb), oxyhemoglobin (HbO), and oxCCO concentrations was investigated. No meaningful differences were found between concentration changes recovered from uncompressed and compressed data, with mean differences of 0.16 ± 0.20 µM, -0.25 ± 0.21 µM, and -0.04 ± 0.10 µM for Hb, HbO, and oxCCO, respectively. The results show that changes in oxCCO and S_tO₂ can be reliably monitored with a high compression rate. Future work will compare the performance of the TR spectrometer with that of a continuous-wave spectrometer to assess accuracy and will investigate the sensitivity of the device to oxCCO and S_tO₂ changes in the bottom compartment of a 2-layer tissue-mimicking phantom.

We demonstrate a novel realization of Interstitial fiber, broadband, Time Domain Diffuse Optical Spectroscopy (TD-DOS) in Null Source-Detector separation (NSDS) approach without temporal gating, by using a Super-conducting Nanowire single photon detector (SNSPD) for acquisition. We test its feasibility by performing Monte Carlo simulations and comparing the absorption retrieval of the SNSPD with an ideal scenario and a standard Silicon Photomultiplier (SiPM). Consequently, as per the MEDPHOT protocol, we test experimentally, the absorption linearity of the system on tissue-equivalent liquid phantoms and demonstrate the scattering independent retrieval of the absorption spectrum of water using Intralipid phantoms in the wavelength range of 600-1100 nm.

Diffuse Correlation Spectroscopy (DCS) provides a non-invasive method of measuring microvasculature cerebral blood flow (CBF). Recent advancements have enabled flow pulsatility to be captured, providing a means of continuously monitoring critical closing pressure (CCP), which is intrinsically linked to intracranial pressure [1]. Similar to DCS measurements of mean CBF, DCS pulsatility data can be contaminated by blood flow in the extracerebral (EC) tissue [2]. This study focuses on extracting CBF pulsatility using the probe pressure modulation algorithm proposed by Baker et al. for removing EC contamination [3]. DCS data were collected from five healthy volunteers, along with continuous recordings of arterial blood pressure and ECG. Data were acquired at two source detector distances (rSD = 1 and 2.5 cm) at a sampling frequency of 20 Hz [4]. The pulsatile waveform was generated by two methods: (1) fitting the semi-infinite model to data acquired at rSD = 2.5 cm and (2) applying the pressure modulation algorithm to data from both distances. The two waveforms were compared based on extracting waveform features, including the systolic-to-diastolic amplitude (YSD), reflective flow peak (S2), dicrotic notch (d), diastolic peak (D), and Δt (Δt = tS1 - td). Preliminary results indicated that removing EC contamination caused a significant increase in Y_SD and Δt. Reductions in S2 and d were also observed, but these changes did not reach statistical significance. In conclusion, these preliminary findings suggest that EC contamination can alter the shape of the pulsatile waveform, which could influence parameters such as CCP used to assess brain health. Collecting multi-distance DCS data and incorporating the pressure modulation algorithm to remove EC contamination is recommended.

Neuromonitoring during cardiac surgery helps prevent brain injury by detecting evidence of cerebral ischemia. Current neuromonitoring devices, such as cerebral oximeters, generally monitor one brain region, which prevents the detection of blood flow impairment in other vascular territories. A potential solution is to use a device with full-head coverage such as the newly developed high-density time-resolved NIRS system, Kernel Flow. This work aimed to assess Kernel Flow’s sensitivity to regional cerebral oxygenation changes using momentary carotid compression (CC), a paradigm that causes substantial decreases in cerebral blood flow and oxyhemoglobin (HbO) throughout the ipsilateral hemisphere. Five healthy volunteers were imaged using a Kernel Flow headset during a 30-s CC. To assess the sensitivity of the device to regional changes, the number of good quality channels was compared between four brain regions: frontal, somatosensory, temporal, and occipital. HbO and deoxyhemoglobin (HbR) time series in the ipsilateral and contralateral hemispheres were analyzed. Overall, the frontal region had the largest amount of good-quality channels, and the ipsilateral regions showed the expected HbO decrease during CC. All contralateral regions showed minimal changes during CC, as expected. Overall, the Flow device showed good sensitivity to reduced cerebral blood flow; however, its use as a neuromonitor during cardiac surgery could be challenged by signal degradation due to hair, although this may be less of an issue with cardiac patients considering that most are older and have less hair.

Cerebral hemodynamics, measured with near-infrared spectroscopy, that are coherent with changes in blood pressure can be analyzed with Coherent Hemodynamics Spectroscopy (CHS). Performing diffuse optical imaging during a CHS protocol provides the ability to spatially map cerebral hemodynamics and elucidate their relation to blood flow and blood volume dynamics. Here, we apply frequency-domain dual-slope optical imaging during a CHS protocol to demonstrate the preferential sensitivity to cerebral hemodynamics of dual-slope frequency-domain measurements as compared to traditional single-distance intensity measurements. Specifically, the results show that dual-slope phase measurements recorded hemodynamics that are mostly associated with blood-flow oscillations (as expected in the brain), while single distance intensity measurements recorded hemodynamics that are mostly associated with blood-volume oscillations (as expected in the scalp). Reconstructed dual-slope phase images showed the effect of a spatially variable skull thickness, which can cause heterogeneity within the reconstructed images. Future work will include measurements on multiple subjects and across multiple oscillation frequencies to further investigate the spatial distribution and frequency dependence of cerebral hemodynamic oscillations.

Intraventricular hemorrhage (IVH) is a common occurrence in preterm infants born with very low birth weights, often leading to hydrocephalus. Hydrocephalus is an abnormal accumulation of cerebral spinal fluid (CSF) in the brain that can cause high intracranial pressure (ICP) and subsequent brain injuries. Unfortunately, current neuromonitoring techniques, such as ultrasonography, can only detect injuries that have already occurred. This emphasizes the need for tools that can identify indicators of brain insult prior to the injury. This study aimed to investigate whether cerebral blood flow (CBF) and oxygenation are sensitive to elevated ICP. Experiments were conducted on five newborn piglets, comprised by an experimental (n = 3) and a control group (n = 2). ICP was increased in the experimental group by continuous infusion of saline into the lumbar CSF region. CBF, deoxy- and oxy-hemoglobin (Hb and HbO₂) were continuously measured, starting 10 min before infusion and throughout the saline infusion, using a hybrid optical device that combines continuous-wave hyperspectral near infrared spectroscopy (h-NIRS) and diffuse correlation spectroscopy (DCS). Changes in CBF, Hb, and HbO₂ were computed using methods reported in our previous works. The results revealed that when ICP increases, Hb increases while HbO₂ and CBF decrease. Notably, there was a strong positive correlation between Hb and ICP and a negative correlation between HbO₂, CBF, and ICP (p<0.05). These findings suggest that CBF, Hb, and HbO₂ are sensitive to increased ICP and could be used to detect hydrocephalus-induced high ICP.

A machine learning classification algorithm is applied to the SOLUS database to discriminate benign and malignant breast lesions, based on absorption and composition properties retrieved through diffuse optical tomography. The Mann-Whitney test indicates oxy-hemoglobin (p-value = 0.0007) and lipids (0.0387) as the most significant constituents for lesion classification, but work is in progress for further analysis. Together with sensitivity (91%), specificity (75%) and the Area Under the ROC Curve (0.83), special metrics for imbalanced datasets (27% of malignant lesions) are applied to the machine learning outcome: balanced accuracy (83%) and Matthews Correlation Coefficient (0.65). The initial results underline the promising informative content of optical data.

The purpose of this clinical study is to monitor NeoAdjuvant Chemotherapy through time domain Diffuse Optical Spectroscopy, correlate the optical results with conventional imaging techniques and pathological response and eventually predict the efficacy of NAC in breast cancer patients. Our seven wavelength (635 -1060 nm) optical mammograph is used to perform non-invasive measurements on patients undergoing NAC in this study. The broad spectral range helps us to fully analyze tissue composition, that includes hemoglobin, water lipids and collagen concentration, to track the tumor response during the course of the therapy. In this paper, we present the preliminary results of five patients.

Thermal therapy is a minimally invasive technique with great potential for treating tumors in non-surgical candidates. Minimally invasive procedures reduce post-operation complications and improve patient quality of life.¹ Thermal therapies work by inducing damage to a tissue using heat. There are several types of Thermal therapies, such as microwave ablation, laser ablation, or radiofrequency ablation. The current limitation is the treatment control² to limit the damage to healthy tissue and treat the cancerous regions entirely. The search for an effective tool for monitoring the thermal outcome is still ongoing. Diffuse optics could monitor during and after the thermal therapy as the optical properties are linked to the damage level³⁴ received by the tissue. In this work, we aimed to study the change in the optical properties of a porcine muscle during a thermal treatment to identify the reactions that occur during the therapy to study their evolution.

By exploiting the recent components miniaturization trend, we realize a small and cheap multifunctional time-resolved (TR) single-photon detection chain. It is based on 16 channels, which can be configured either as 16 independently located channels for TR diffuse optical tomography or as a linear array for parallel TR fluorescence spectroscopy. Both applications require a detector with high time resolution and high light harvesting capability (i.e., large active area and detection efficiency). Thus, each detection channel contains a 1.3 x 1.3 mm2 active area silicon photomultiplier and its home-made electronics specifically designed for avalanche sensing and amplification, capable of optimizing the single-photon timing resolution despite the miniaturization. In this study we describe the timing performances of a first 8- channel prototype and its first application in fluorescence lifetime sensing. Then, we show the capability of the whole 16-channel array in detecting absorption changes within a homogeneous scattering medium. We have been able to obtain a single-photon timing resolution of almost 60 ps, that is close to the best ever achieved with this kind of detector. For the validation in fluorescence lifetime sensing, the fluorescence signal acquired by the proposed prototype is comparable to the one obtained using a state-of-the-art setup based on a PMT detector. In the validation in diffuse optics, we clearly detected the absorption perturbation. This confirms the suitability of this stackable solid-state detector array for both applications.

The invention of the random laser has opened a new frontier in optics, providing also the opportunity to explore new possibilities in the field of sensing. The research in optical sensors has indeed been largely encouraged by the demand for low-cost and non-invasive new detection strategies. The main advantage in exploiting the physical principle of the random laser in optical sensors is due to the presence of the stimulated emission mechanism, which allows amplification and spectral modification of the signal. We present here a step forward in the exploitation of this optical sensor device by an improved revisitation of a previous experimental setup, both in the instrumentation and in the measurement method, to mitigate the instability of the results due to shot-to-shot pump energy fluctuations. The novelties introduced, the use of optical fibers, a reference sensor, and a peristaltic pump have shown to eliminate optical beam alignment issues and the problems linked to variation in pump energy. The implemented sensor allows easy and rapid change of the sensed medium. These results pave the way for a portable device to directly measure the scattering of liquid samples, without resorting to complicated numerical or analytic inversion procedures of the measured data, provided that a suitable calibration of the system is performed.

Traditionally, in biomedical optics, the photons mean fluence rate assessed in a sub-volume of a propagating medium is obtained with Monte Carlo (MC) simulations by calculating the deposited power by the absorbed photons in the sub-volume. We propose an alternative method based on the assessment of the mean pathlength traveled by all the injected photons inside the sub-volume. Examples of its applications are given. This method also works for nil absorption coefficient and for a non-constant spatial distribution of the absorption coefficient inside the sub-volume. The proposed approach is a re-visitation of a well-known method applied in radiation and nuclear physics. The relation at the basis of the method descends from the ground definitions of quantities employed in radiative transfer. The results obtained show that a potential advantage of the proposed method is that it can improve the convergence of the MC simulations. Indeed, when calculating the fluence in a region of interest with the proposed method all the photons that cross the region are considered. While, with the traditional approach only the “absorbed”photons can contribute to the calculated fluence. In the latter case, this may produce a poorer MC statistic for the same number of launched photons.

The Monte Carlo method is a gold standard for “solving” the radiative transport equation even in complex geometries and distributions of the optical properties. The exact analytical benchmark provided by the law of the invariant total mean pathlength spent by light injected with uniform Lambertian illumination inside non-absorbing scattering media is used to verify Monte Carlo codes developed for biomedical optics applications. The correctness of an MC code can be assessed with a one-sample t-test. Further, the invariance of the average path length guarantees that the expected value is known regardless of the complexity of the medium. The results obtained show that the accuracy of the estimated average pathlength can be progressively increase as the number of simulated trajectories increases. The method can be applied in total generality versus the scattering and geometrical properties of the medium, as well as in presence of refractive index mismatch between the medium and the external region and between different regions of the medium. The proposed verification method is especially reliable to detect inaccuracies in the treatment of boundaries of finite media. The results presented in this contribution, obtained by a standard computer machine, show a verification of our Monte Carlo code up to the sixth decimal digit. This method can provide a fundamental tool for the verification of Monte Carlo codes in the geometry of interest, without resorting to simpler geometries and uniform distribution of the scattering properties.

We present line scan reflectance diffuse optical tomography (LS-RDOT), a technique to generate quantitative cross-sectional images of hemoglobin concentration, tissue oxygen saturation, water content, and lipid content, for non-invasive bedside imaging of breast cancer. The LS-RDOT system is composed of a single-channel time-domain diffuse optical spectroscopy (TD-DOS) system measuring at wavelengths of 761, 802, 838, 908, 936, and 976 nm and hand-held probes with source–detector distances of 20, 30, and 40 mm. The line scans were performed by acquiring temporal point spread functions (TPSF) at 9 measurement points with a spacing of 5 mm linearly marked on the skin just above the breast lesion. The cross-sectional images were restored by an iterative image reconstruction method with an expression of the TPSF obtained from the photon diffusion equation using the Rytov approximation. A preliminary clinical measurement was conducted for a breast cancer patient with a tumor of approximately 10 mm in size. The reconstructed images captured changes in the physiological parameters of the breast cancer at the lesion location indicated by the ultrasonographic image. In addition, the results showed that LS-RDOT provides cross-sectional images of physiological parameters in a form that can be fused with structural images provided by ultrasonography