Next-generation acceleration and code optimization for light transport in turbid media using GPUs

Test-time augmentation with synthetic data addresses distribution shifts in spectral imaging

PURPOSE

Surgical scene segmentation is crucial for providing context-aware surgical assistance. Recent studies highlight the significant advantages of hyperspectral imaging (HSI) over traditional RGB data in enhancing segmentation performance. Nevertheless, the current hyperspectral imaging (HSI) datasets remain limited and do not capture the full range of tissue variations encountered clinically.

METHODS

Based on a total of 615 hyperspectral images from a total of 16 pigs, featuring porcine organs in different perfusion states, we carry out an exploration of distribution shifts in spectral imaging caused by perfusion alterations. We further introduce a novel strategy to mitigate such distribution shifts, utilizing synthetic data for test-time augmentation.

RESULTS

The effect of perfusion changes on state-of-the-art (SOA) segmentation networks depended on the organ and the specific perfusion alteration induced. In the case of the kidney, we observed a performance decline of up to 93% when applying a state-of-the-art (SOA) network under ischemic conditions. Our method improved on the state-of-the-art (SOA) by up to 4.6 times.

CONCLUSION

Given its potential wide-ranging relevance to diverse pathologies, our approach may serve as a pivotal tool to enhance neural network generalization within the realm of spectral imaging.

Diffuse reflectance spectroscopy for colorectal cancer surgical guidance: towards real-time tissue characterization and new biomarkers

Colorectal cancer (CRC) is the third most common and second most deadly type of cancer worldwide, representing 11.3% of the diagnosed cancer cases and resulting in 10.2% (0.88 million) of the cancer related deaths in 2020. CRCs are typically detected at the late stage, which leads to high mortality and morbidity. Mortality and poor prognosis are partially caused by cancer recurrence and postoperative complications. Patient survival could be increased by improving precision in surgical resection using accurate surgical guidance tools based on diffuse reflectance spectroscopy (DRS). DRS enables real-time tissue identification for potential cancer margin delineation through determination of the circumferential resection margin (CRM), while also supporting non-invasive and label-free approaches for laparoscopic surgery to avoid short-term complications of open surgery as suitable. In this study, we have estimated the scattering properties and chromophore concentrations based on 2949 DRS measurements of freshly excised ex vivo specimens of 47 patients, and used this estimation to classify normal colorectal wall (CW), fat and tumor tissues. DRS measurements were performed with fiber-optic probes of 630 μm source-detector distance (SDD; probe 1) and 2500 μm SDD (probe 2) to measure tissue layers ∼0.5-1 mm and ∼0.5-2 mm deep, respectively. By using the 5-fold cross-validation of machine learning models generated with the classification and regression tree (CART) algorithm, we achieved 95.9 ± 0.7% sensitivity, 98.9 ± 0.3% specificity, 90.2 ± 0.4% accuracy, and 95.5 ± 0.3% AUC for probe 1. Similarly, we achieved 96.9 ± 0.8% sensitivity, 98.9 ± 0.2% specificity, 94.0 ± 0.4% accuracy, and 96.7 ± 0.4% AUC for probe 2.

Spectral imaging enables contrast agent-free real-time ischemia monitoring in laparoscopic surgery

Laparoscopic surgery has evolved as a key technique for cancer diagnosis and therapy. While characterization of the tissue perfusion is crucial in various procedures, such as partial nephrectomy, doing so by means of visual inspection remains highly challenging. We developed a laparoscopic real-time multispectral imaging system featuring a compact and lightweight multispectral camera and the possibility to complement the conventional surgical view of the patient with functional information at a video rate of 25 Hz. To enable contrast agent-free ischemia monitoring during laparoscopic partial nephrectomy, we phrase the problem of ischemia detection as an out-of-distribution detection problem that does not rely on data from any other patient and uses an ensemble of invertible neural networks at its core. An in-human trial demonstrates the feasibility of our approach and highlights the potential of spectral imaging combined with advanced deep learning-based analysis tools for fast, efficient, reliable, and safe functional laparoscopic imaging.

Continuous Sizing and Identification of Microplastics in Water

The pollution of the environment with microplastics in general, and in particular, the contamination of our drinking water and other food items, has increasingly become the focus of public attention in recent years. In order to better understand the entry pathways into the human food chain and thus prevent them if possible, a precise characterization of the particles concerning their size and material is indispensable. Particularly small plastic particles pose a special challenge since their material can only be determined by means of large experimental effort. In this work, we present a proof of principle experiment that allows the precise determination of the plastic type and the particle size in a single step. The experiment combines elastic light scattering (Mie scattering) with inelastic light scattering (Raman scattering), the latter being used to determine the plastic type. We conducted Monte Carlo simluations for the elastically scattered light for different kinds of plastics in a microfluidic cuvette which we could reproduce in the experiment. We were able to measure the Raman signals for different microplastics in the same measurement as the elastically scattered light and thereby determine their material. This information was used to select the appropriate Monte Carlo simulation data and to assign the correct particle size to different materials with only one calibration measurement.

A Monte Carlo simulation for Moving Light Source in Intracavity PDT

We developed a simulation method for modeling the light fluence delivery in intracavity Photodynamic Therapy (icav-PDT) for pleural lung cancer using a moving light source. Due to the large surface area of the pleural lung cavity, the light source needs to be moved to deliver a uniform dose around the entire cavity. While multiple fixed detectors are used for dosimetry at a few locations, an accurate simulation of light fluence and fluence rate is still needed for the rest of the cavity. We extended an existing Monte Carlo (MC) based light propagation solver to support moving light sources by densely sampling the continuous light source trajectory and assigning the proper number of photon packages launched along the way. The performance of Simphotek GPU CUDA-based implementation of the method - PEDSy-MC - has been demonstrated on a life-size lung-shaped phantom, custom printed for testing icav-PDT navigation system at the Perlman School of Medicine (PSM) - calculations completed under a minute (for some cases) and within minutes have been achieved. We demonstrate results within a 5% error of the analytic solution for multiple detectors in the phantom. PEDSy-MC is accompanied by a dose-cavity visualization tool that allows real-time inspection of dose values of the treated cavity in 2D and 3D, which will be expanded to ongoing clinical trials at PSM. PSM has developed a technology to measure 8-detectors in a pleural cavity phantom using Photofrin-mediated PDT that has been used during validation.

Lawrence K, Tachtsidis I, Tong Y, Torricelli A, Urner T, Wabnitz H, Wolf M, Wolf U, Xu S, Yang C, Yodh AG, Yücel MA, Zhou W. Optical imaging and spectroscopy for the study of the human brain: status report

This report is the second part of a comprehensive two-part series aimed at reviewing an extensive and diverse toolkit of novel methods to explore brain health and function. While the first report focused on neurophotonic tools mostly applicable to animal studies, here, we highlight optical spectroscopy and imaging methods relevant to noninvasive human brain studies. We outline current state-of-the-art technologies and software advances, explore the most recent impact of these technologies on neuroscience and clinical applications, identify the areas where innovation is needed, and provide an outlook for the future directions.

Convolutional neural network-based common-path optical coherence tomography A-scan boundary-tracking training and validation using a parallel Monte Carlo synthetic dataset

We present a parallel Monte Carlo (MC) simulation platform for rapidly generating synthetic common-path optical coherence tomography (CP-OCT) A-scan image dataset for image-guided needle insertion. The computation time of the method has been evaluated on different configurations and 100000 A-scan images are generated based on 50 different eye models. The synthetic dataset is used to train an end-to-end convolutional neural network (Ascan-Net) to localize the Descemet's membrane (DM) during the needle insertion. The trained Ascan-Net has been tested on the A-scan images collected from the ex-vivo human and porcine cornea as well as simulated data and shows improved tracking accuracy compared to the result by using the Canny-edge detector.

Band selection for oxygenation estimation with multispectral/hyperspectral imaging

Multispectral imaging provides valuable information on tissue composition such as hemoglobin oxygen saturation. However, the real-time application of this technique in interventional medicine can be challenging due to the long acquisition times needed for large amounts of hyperspectral data with hundreds of bands. While this challenge can partially be addressed by choosing a discriminative subset of bands, the band selection methods proposed to date are mainly restricted by the availability of often hard to obtain reference measurements. We address this bottleneck with a new approach to band selection that leverages highly accurate Monte Carlo (MC) simulations. We hypothesize that a so chosen small subset of bands can reproduce or even improve upon the results of a quasi continuous spectral measurement. We further investigate whether novel domain adaptation techniques can address the inevitable domain shift stemming from the use of simulations. Initial results based on in silico and in vivo experiments suggest that 10-20 bands are sufficient to closely reproduce results from spectral measurements with 101 bands in the 500-700 nm range. The investigated domain adaptation technique, which only requires unlabeled in vivo measurements, yielded better results than the pure in silico band selection method. Overall, our method could guide development of fast multispectral imaging systems suited for interventional use without relying on complex hardware setups or manually labeled data.

Multi-scattering software part II: experimental validation for the light intensity distribution

This article, Part II of an article series on GPU-accelerated Monte Carlo simulation of photon transport through turbid media, focuses on the validation of the online software Multi-Scattering. While Part I detailed the implementation of the computational model, simulated and experimental results are now compared for the distribution of the scattered light intensity. The scattering phantoms prepared here are aqueous dispersions of polystyrene microspheres of diameter D = 0.5, 2 and 5 μm and at various concentrations, resulting in optical depth ranging from OD = 1 to 17.5. The Lorenz-Mie scattering phase functions used in the simulations have been verified experimentally at low particle concentrations by analyzing the angular light intensity distribution at the Fourier plane of a collecting lens. The validation approach herein accounts for the specific light collection and image formation by the camera. The front and side surfaces of the medium are imaged and the corresponding light intensity distributions are compared qualitatively and quantitatively. It is concluded that the model enables reliable simulations over the tested parameters, offering predictive simulations of transmitted intensities with a mean relative error ≤~19% over the full range. The online software is available at: https://multi-scattering.com/.

Imaging sub-diffuse optical properties of cancerous and normal skin tissue using machine learning-aided spatial frequency domain imaging

SIGNIFICANCE

Sub-diffuse optical properties may serve as useful cancer biomarkers, and wide-field heatmaps of these properties could aid physicians in identifying cancerous tissue. Sub-diffuse spatial frequency domain imaging (sd-SFDI) can reveal such wide-field maps, but the current time cost of experimentally validated methods for rendering these heatmaps precludes this technology from potential real-time applications.

AIM

Our study renders heatmaps of sub-diffuse optical properties from experimental sd-SFDI images in real time and reports these properties for cancerous and normal skin tissue subtypes.

APPROACH

A phase function sampling method was used to simulate sd-SFDI spectra over a wide range of optical properties. A machine learning model trained on these simulations and tested on tissue phantoms was used to render sub-diffuse optical property heatmaps from sd-SFDI images of cancerous and normal skin tissue.

RESULTS

The model accurately rendered heatmaps from experimental sd-SFDI images in real time. In addition, heatmaps of a small number of tissue samples are presented to inform hypotheses on sub-diffuse optical property differences across skin tissue subtypes.

CONCLUSION

These results bring the overall process of sd-SFDI a fundamental step closer to real-time speeds and set a foundation for future real-time medical applications of sd-SFDI such as image guided surgery.

In-silico investigation towards the non-invasive optical detection of blood lactate

This paper uses Monte Carlo simulations to investigate the interaction of short-wave infrared (SWIR) light with vascular tissue as a step toward the development of a non-invasive optical sensor for measuring blood lactate in humans. The primary focus of this work was to determine the optimal source-detector separation, penetration depth of light at SWIR wavelengths in tissue, and the optimal light power required for reliable detection of lactate. The investigation also focused on determining the non-linear variations in absorbance of lactate at a few select SWIR wavelengths. SWIR photons only penetrated 1.3 mm and did not travel beyond the hypodermal fat layer. The maximum output power was only 2.51% of the input power, demonstrating the need for a highly sensitive detection system. Simulations optimized a source-detector separation of 1 mm at 1684 nm for accurate measurement of lactate in blood.

Spatially-Resolved Multiply-Excited Autofluorescence and Diffuse Reflectance Spectroscopy: SpectroLive Medical Device for Skin In Vivo Optical Biopsy

This contribution presents the development of an optical spectroscopy device, called SpectroLive, that allows spatially-resolved multiply-excited autofluorescence and diffuse reflectance measurements. Besides describing the device, this study aims at presenting the metrological and safety regulation validations performed towards its aimed application to skin carcinoma in vivo diagnosis. This device is made of six light sources and four spectrometers for detection of the back-scattered intensity spectra collected through an optical probe (made of several optical fibers) featuring four source-to-detector separations (from 400 to 1000 µm). In order to be allowed by the French authorities to be evaluated in clinics, the SpectroLive device was successfully checked for electromagnetic compatibility and electrical and photobiological safety. In order to process spectra, spectral correction and metrological calibration were implemented in the post-processing software. Finally, we characterized the device’s sensitivity to autofluorescence detection: excitation light irradiance at the optical probe tip in contact with skin surface ranges from 2 to 11 W/m², depending on the light source. Such irradiances combined to sensitive detectors allow the device to acquire a full spectroscopic sequence within 6 s which is a short enough duration to be compatible with optical-guided surgery. All these results about sensitivity and safety make the SpectroLive device mature enough to be evaluated through a clinical trial that aims at evaluating its diagnostic accuracy for skin carcinoma diagnosis.

Evaluation of wavelength ranges and tissue depth probed by diffuse reflectance spectroscopy for colorectal cancer detection

Colorectal cancer (CRC) is the third most common type of cancer worldwide and the second most deadly. Recent research efforts have focused on developing non-invasive techniques for CRC detection. In this study, we evaluated the diagnostic capabilities of diffuse reflectance spectroscopy (DRS) for CRC detection by building 6 classification models based on support vector machines (SVMs). Our dataset consists of 2889 diffuse reflectance spectra collected from freshly excised ex vivo tissues of 47 patients over wavelengths ranging from 350 and 1919 nm with source-detector distances of 630-µm and 2500-µm to probe different depths. Quadratic SVMs were used and performance was evaluated using twofold cross-validation on 10 iterations of randomized training and test sets. We achieved (93.5 ± 2.4)% sensitivity, (94.0 ± 1.7)% specificity AUC by probing the superficial colorectal tissue and (96.1 ± 1.8)% sensitivity, (95.7 ± 0.6)% specificity AUC by sampling deeper tissue layers. To the best of our knowledge, this is the first DRS study to investigate the potential of probing deeper tissue layers using larger SDD probes for CRC detection in the luminal wall. The data analysis showed that using a broader spectrum and longer near-infrared wavelengths can improve the diagnostic accuracy of CRC as well as probing deeper tissue layers.

Noninvasive Monitoring of Dynamical Processes in Bruised Human Skin Using Diffuse Reflectance Spectroscopy and Pulsed Photothermal Radiometry

We have augmented a recently introduced method for noninvasive analysis of skin structure and composition and applied it to monitoring of dynamical processes in traumatic bruises. The approach combines diffuse reflectance spectroscopy in visible spectral range and pulsed photothermal radiometry. Data from both techniques are analyzed simultaneously using a numerical model of light and heat transport in a four-layer model of human skin. Compared to the earlier presented approach, the newly introduced elements include two additional chromophores (β-carotene and bilirubin), individually adjusted thickness of the papillary dermal layer, and analysis of the bruised site using baseline values assessed from intact skin in its vicinity. Analyses of traumatic bruises in three volunteers over a period of 16 days clearly indicate a gradual, yet substantial increase of the dermal blood content and reduction of its oxygenation level in the first days after injury. This is followed by the emergence of bilirubin and relaxation of all model parameters towards the values characteristic for healthy skin approximately two weeks after the injury. The assessed parameter values and time dependences are consistent with existing literature. Thus, the presented methodology offers a viable approach for objective characterization of the bruise healing process.

Multi-Scattering software: part I: online accelerated Monte Carlo simulation of light transport through scattering media

In this article we present and describe an online freely accessible software called Multi-Scattering for the modeling of light propagation in scattering and absorbing media. Part II of this article series focuses on the validation of the model by rigorously comparing the simulated results with experimental data. The model is based on the use of the Monte Carlo method, where billions of photon packets are being tracked through simulated cubic volumes. Simulations are accelerated by the use of general-purpose computing on graphics processing units, reducing the computation time by a factor up to 200x in comparison with a single central processing unit thread. By using four graphic cards on a single computer, the simulation speed increases by a factor of 800x. For an anisotropy factor g = 0.86, this enables the transport path of one billion photons to be computed in 10 seconds for optical depth OD = 10 and in 20 minutes for OD = 500. Another feature of Multi-Scattering is the integration and implementation of the Lorenz-Mie theory in the software to generate the scattering phase functions from spherical particles. The simulations are run from a computer server at Lund University, allowing researchers to log in and use it freely without any prior need for programming skills or specific software/hardware installations. There are countless types of scattering media in which this model can be used to predict light transport, including medical tissues, blood samples, clouds, smoke, fog, turbid liquids, spray systems, etc. An example of simulation results is given here for photon propagation through a piece of human head. The software also includes features for modeling image formation by inserting a virtual collecting lens and a detection matrix which simulate a camera objective and a sensor array respectively. The user interface for setting-up simulations and for displaying the corresponding results is found at: https://multi-scattering.com/.

Exploiting scale-invariance: a top layer targeted inverse model for hyperspectral images of wounds

Detection of re-epithelialization in wound healing is important, but challenging. Hyperspectral imaging can be used for non-destructive characterization, but efficient techniques are needed to extract and interpret the information. An inverse photon transport model suitable for characterization of re-epithelialization is validated and explored in this study. It exploits scale-invariance to enable fitting of the epidermal skin layer only. Monte Carlo simulations indicate that the fitted layer transmittance and reflectance spectra are unique, and that there exists an infinite number of coupled parameter solutions. The method is used to explain the optical behavior of and detect re-epithelialization in an in vitro wound model.

Cranial Perforation Using an Optically-Enhanced Surgical Drill

The design of mechanically clutched cranial perforators, used in craniotomy procedures, limits their performance under certain clinical conditions and can, in some cases, impose the risk of severe brain injury on patients undergoing the procedure. An additional safety mechanism could help in mitigating these risks. In this work, we examine the use of diffuse reflectance spectroscopy as a potential fallback mechanism for near real-time detection of the bone-brain boundary. Monte Carlo simulation of a two layer model with optical properties of bone and brain at 530 and 850 nm resulted in a detectable change in diffuse reflectance signal when approaching the boundary. The simulated results were used to guide the development of an experimental drill control system, which was tested on 10 sheep craniums and yielded 88.1 % success rate in the detection of the approaching bone-brain boundary.

Precise determination of the optical properties of turbid media using an optimized integrating sphere and advanced Monte Carlo simulations. Part 1: theory

In this paper, we describe a method used to determine the optical properties, namely, the effective scattering and absorption coefficients, employing an optimized three-dimensional-printed single integrating sphere. The paper consists of two parts, and in Part 1, the theoretical investigation of an optimized measurement and the evaluation routine are presented. Using an analytical and a numerical model for the optical characterization of the integrating sphere, errors caused by the application of a non-ideal sphere (the one with ports or baffles) were investigated. Considering this research, a procedure for the precise determination of the optical properties, based on Monte Carlo simulations of the light distribution within the sample, was developed. In Part 2, we present the experimental validation of this procedure.

Predictive model for the quantitative analysis of human skin using photothermal radiometry and diffuse reflectance spectroscopy

We have recently introduced a novel methodology for the noninvasive analysis of the structure and composition of human skin in vivo. The approach combines pulsed photothermal radiometry (PPTR), involving time-resolved measurements of mid-infrared emission after irradiation with a millisecond light pulse, and diffuse reflectance spectroscopy (DRS) in the visible part of the spectrum. Simultaneous fitting of both data sets with respective predictions from a numerical model of light transport in human skin enables the assessment of the contents of skin chromophores (melanin, oxy-, and deoxy-hemoglobin), as well as scattering properties and thicknesses of the epidermis and dermis. However, the involved iterative optimization of 14 skin model parameters using a numerical forward model (i.e., inverse Monte Carlo - IMC) is computationally very expensive. In order to overcome this drawback, we have constructed a very fast predictive model (PM) based on machine learning. The PM involves random forests, trained on ∼9,000 examples computed using our forward MC model. We show that the performance of such a PM is very satisfying, both in objective testing using cross-validation and in direct comparisons with the IMC procedure. We also present a hybrid approach (HA), which combines the speed of the PM with versatility of the IMC procedure. Compared with the latter, the HA improves both the accuracy and robustness of the inverse analysis, while significantly reducing the computation times.

Optical investigation of three-dimensional human skin equivalents: A pilot study

Human skin equivalents (HSEs) are three-dimensional living models of human skin that are prepared in vitro by seeding cells onto an appropriate scaffold. They recreate the structure and biological behaviour of real skin, allowing the investigation of processes such as keratinocyte differentiation and interactions between the dermal and epidermal layers. However, for wider applications, their optical and mechanical properties should also replicate those of real skin. We therefore conducted a pilot study to investigate the optical properties of HSEs. We compared Monte Carlo simulations of (a) real human skin and (b) two-layer optical models of HSEs with (c) experimental measurements of transmittance through HSE samples. The skin layers were described using a hybrid collection of optical attenuation coefficients. A linear relationship was observed between the simulations and experiments. For samples thinner than 0.5 mm, an exponential increase in detected power was observed due to fewer instances of absorption and scattering.

Physiological model using diffuse reflectance spectroscopy for nonmelanoma skin cancer diagnosis

Diffuse reflectance spectroscopy (DRS) is a noninvasive, fast, and low-cost technology with potential to assist cancer diagnosis. The goal of this study was to test the capability of our physiological model, a computational Monte Carlo lookup table inverse model, for nonmelanoma skin cancer diagnosis. We applied this model on a clinical DRS dataset to extract scattering parameters, blood volume fraction, oxygen saturation and vessel radius. We found that the model was able to capture physiological information relevant to skin cancer. We used the extracted parameters to classify (basal cell carcinoma [BCC], squamous cell carcinoma [SCC]) vs actinic keratosis (AK) and (BCC, SCC, AK) vs normal. The area under the receiver operating characteristic curve achieved by the classifiers trained on the parameters extracted using the physiological model is comparable to that of classifiers trained on features extracted via Principal Component Analysis. Our findings suggest that DRS can reveal physiologic characteristics of skin and this physiologic model offers greater flexibility for diagnosing skin cancer than a pure statistical analysis. Physiological parameters extracted from diffuse reflectance spectra data for nonmelanoma skin cancer diagnosis.

Graphics processing unit-accelerated mesh-based Monte Carlo photon transport simulations

The mesh-based Monte Carlo (MMC) algorithm is increasingly used as the gold-standard for developing new biophotonics modeling techniques in 3-D complex tissues, including both diffusion-based and various Monte Carlo (MC)-based methods. Compared to multilayered and voxel-based MCs, MMC can utilize tetrahedral meshes to gain improved anatomical accuracy but also results in higher computational and memory demands. Previous attempts of accelerating MMC using graphics processing units (GPUs) have yielded limited performance improvement and are not publicly available. We report a highly efficient MMC-MMCL-using the OpenCL heterogeneous computing framework and demonstrate a speedup ratio up to 420× compared to state-of-the-art single-threaded CPU simulations. The MMCL simulator supports almost all advanced features found in our widely disseminated MMC software, such as support for a dozen of complex source forms, wide-field detectors, boundary reflection, photon replay, and storing a rich set of detected photon information. Furthermore, this tool supports a wide range of GPUs/CPUs across vendors and is freely available with full source codes and benchmark suites at http://mcx.space/#mmc.

Retrieval of Absorption or Scattering Coefficient Spectrum (RASCS) Program: A Tool to Monitor Optical Properties in Real Time

BACKGROUND AND OBJECTIVES

Optical properties characterize light propagation in turbid media, such as tissue. Recovery of optical properties is of great importance in a wide variety of biomedical applications, including both therapeutic treatments and diagnosis. Most of the available methodologies are well established, however, these are not optimized for real-time measurements.

STUDY DESIGN/MATERIALS AND METHODS

Optical properties are recovered using the Inverse Adding Doubling program from reflectance measurements measured with an integrating sphere and light in the visible range. A user-friendly interface was programmed in Visual Studio and the libraries of a particular spectrophotometer were used. To achieve real-time measurements, a parallel computing routine was implemented, splitting the whole spectra in threads to be computed independently. Several tests using living tissue and inorganic materials were carried out to validate the proposed algorithm.

RESULTS

Recovery of absorption/scattering coefficient spectrum in the visible range with high precision in a couple of seconds was achieved, demonstrating its capabilities for real-time monitoring in biomedical applications. The absorption coefficient spectrum shows the expected characteristics according to the different melanin and blood concentration of various volunteers, also showing the expected changes during a thermoregulation process.

CONCLUSIONS

A real-time monitoring of optical properties algorithm was developed, including parallel computing and a user-friendly interface. The proposed algorithm would be of help in biomedical applications, where real-time monitoring optical properties is required. Lasers Surg. Med. © 2019 Wiley Periodicals, Inc.

FullMonteCUDA: a fast, flexible, and accurate GPU-accelerated Monte Carlo simulator for light propagation in turbid media

Optimizing light delivery for photodynamic therapy, quantifying tissue optical properties or reconstructing 3D distributions of sources in bioluminescence imaging and absorbers in diffuse optical imaging all involve solving an inverse problem. This can require thousands of forward light propagation simulations to determine the parameters to optimize treatment, image tissue or quantify tissue optical properties, which is time-consuming and computationally expensive. Addressing this problem requires a light propagation simulator that produces results quickly given modelling parameters. In previous work, we developed FullMonteSW: currently the fastest, tetrahedral-mesh, Monte Carlo light propagation simulator written in software. Additional software optimizations showed diminishing performance improvements, so we investigated hardware acceleration methods. This work focuses on FullMonteCUDA: a GPU-accelerated version of FullMonteSW which targets NVIDIA GPUs. FullMonteCUDA has been validated across several benchmark models and, through various GPU-specific optimizations, achieves a 288-936x speedup over the single-threaded, non-vectorized version of FullMonteSW and a 4-13x speedup over the highly optimized, hand-vectorized and multi-threaded version. The increase in performance allows inverse problems to be solved more efficiently and effectively.

Advances in the simulation of light-tissue interactions in biomedical engineering

Monte Carlo (MC) simulation for light propagation in scattering and absorbing media is the gold standard for studying the interaction of light with biological tissue and has been used for years in a wide variety of cases. The interaction of photons with the medium is simulated based on its optical properties and the original approximation of the scattering phase function. Over the past decade, with the new measurement geometries and recording techniques invented also the corresponding sophisticated methods for the description of the underlying light-tissue interaction taking into account realistic parameters and settings were developed. Applications, such as multiple scattering, optogenetics, optical coherence tomography, Raman spectroscopy, polarimetry and Mueller matrix measurement have emerged and are still constantly improved. Here, we review the advances and recent applications of MC simulation for the active field of the life sciences and the medicine pointing out the new insights enabled by the theoretical concepts.

Raman Sensing and Its Multimodal Combination with Optoacoustics and OCT for Applications in the Life Sciences

Currently, many optical modalities are being investigated, applied, and further developed for non-invasive analysis and sensing in the life sciences. To befit the complexity of the study objects and questions in this field, the combination of two or more modalities is attempted. We review our work on multimodal sensing concepts for applications ranging from non-invasive quantification of biomolecules in the living organism to supporting medical diagnosis showing the combined capabilities of Raman spectroscopy, optical coherence tomography, and optoacoustics.

Uncertainty-aware performance assessment of optical imaging modalities with invertible neural networks

PURPOSE

Optical imaging is evolving as a key technique for advanced sensing in the operating room. Recent research has shown that machine learning algorithms can be used to address the inverse problem of converting pixel-wise multispectral reflectance measurements to underlying tissue parameters, such as oxygenation. Assessment of the specific hardware used in conjunction with such algorithms, however, has not properly addressed the possibility that the problem may be ill-posed.

METHODS

We present a novel approach to the assessment of optical imaging modalities, which is sensitive to the different types of uncertainties that may occur when inferring tissue parameters. Based on the concept of invertible neural networks, our framework goes beyond point estimates and maps each multispectral measurement to a full posterior probability distribution which is capable of representing ambiguity in the solution via multiple modes. Performance metrics for a hardware setup can then be computed from the characteristics of the posteriors.

RESULTS

Application of the assessment framework to the specific use case of camera selection for physiological parameter estimation yields the following insights: (1) estimation of tissue oxygenation from multispectral images is a well-posed problem, while (2) blood volume fraction may not be recovered without ambiguity. (3) In general, ambiguity may be reduced by increasing the number of spectral bands in the camera.

CONCLUSION

Our method could help to optimize optical camera design in an application-specific manner.

Dual-grid mesh-based Monte Carlo algorithm for efficient photon transport simulations in complex three-dimensional media

The mesh-based Monte Carlo (MMC) method is an efficient algorithm to model light propagation inside tissues with complex boundaries, but choosing appropriate mesh density can be challenging. A fine mesh improves the spatial resolution of the output but requires more computation. We propose an improved MMC-dual-grid mesh-based Monte Carlo (DMMC)-to accelerate photon simulations using a coarsely tessellated tetrahedral mesh for ray-tracing computation and an independent voxelated grid for output data storage. The decoupling between ray-tracing and data storage grids allows us to simultaneously achieve faster simulations and improved output spatial accuracy. Furthermore, we developed an optimized ray-tracing technique to eliminate unnecessary ray-tetrahedron intersection tests in optically thick mesh elements. We validate the proposed algorithms using a complex heterogeneous domain and compare the solutions with those from MMC and voxel-based Monte Carlo. We found that DMMC with an unrefined constrained Delaunay tessellation of the boundary nodes yielded the highest speedup, ranging from 1.3  ×   to 2.9  ×   for various scattering settings, with nearly no loss in accuracy. In addition, the optimized ray-tracing technique offers excellent acceleration in high-scattering media, reducing the ray-tetrahedron test count by over 100-fold. Our DMMC software can be downloaded at http://mcx.space/mmc.

Comparison of time- and angular-domain scatter rejection in mesoscopic optical projection tomography: a simulation study

Optical imaging offers exquisite sensitivity and resolution for assessing biological tissue in microscopy applications; however, for samples that are greater than a few hundred microns in thickness (such as whole tissue biopsies), spatial resolution is substantially limited by the effects of light scattering. To improve resolution, time- and angular-domain methods have been developed to reject detection of highly scattered light. This work utilizes a modified version of a commonly used Monte Carlo light propagation software package (MCML) to present the first comparison of time- and angular-domain improvements in spatial resolution with respect to varying sample thickness and optical properties (absorption and scattering). Specific comparisons were made at various tissue thicknesses (1-6 mm) assuming either typical (average) soft tissue scattering properties, μ_s ' = 10 cm^-1, or low scattering properties, μ_s ' = 3.4 cm^-1, as measured in lymph nodes.

Metropolis Monte Carlo simulation scheme for fast scattered X-ray photon calculation in CT

Monte Carlo (MC) method is commonly considered as the most accurate approach for particle transport simulation because of its capability to precisely model physics interactions and simulation geometry. Conventionally, MC simulation is performed in a particle-by-particle fashion. In certain problems such as computing scattered X-ray photon signal at a detector of CT, the conventional simulation scheme suffers from low efficiency mainly due to the fact that abundant photons are simulated but do not reach the detector. The computational resources spent on those photons are therefore wasted. To solve this problem, this study develops a novel GPU-based Metropolis MC (gMMC) with a novel path-by-path simulation scheme and demonstrates its effectiveness in an example problem of scattered X-ray photon calculation in CT. In contrast to the conventional MC approach, gMMC samples an entire photon path extending from the X-ray source to the detector using Metropolis-Hasting algorithm. The path-by-path simulation scheme ensures contribution of every sampled event to the signal of interest, improving overall efficiency. We benchmark gMMC against an in-house developed GPU-based MC tool, gMCDRR, which performs simulations in the conventional particle-by-particle fashion. gMMC reaches speed up factors of 37~48 times in simple phantom cases and 20-34 times in real patient cases. The results calculated by gMCDRR and gMMC agree well with average differences < 3%.

Graphics processing units-accelerated adaptive nonlocal means filter for denoising three-dimensional Monte Carlo photon transport simulations

The Monte Carlo (MC) method is widely recognized as the gold standard for modeling light propagation inside turbid media. Due to the stochastic nature of this method, MC simulations suffer from inherent stochastic noise. Launching large numbers of photons can reduce noise but results in significantly greater computation times, even with graphics processing units (GPU)-based acceleration. We develop a GPU-accelerated adaptive nonlocal means (ANLM) filter to denoise MC simulation outputs. This filter can effectively suppress the spatially varying stochastic noise present in low-photon MC simulations and improve the image signal-to-noise ratio (SNR) by over 5 dB. This is equivalent to the SNR improvement of running nearly 3.5  ×   more photons. We validate this denoising approach using both homogeneous and heterogeneous domains at various photon counts. The ability to preserve rapid optical fluence changes is also demonstrated using domains with inclusions. We demonstrate that this GPU-ANLM filter can shorten simulation runtimes in most photon counts and domain settings even combined with our highly accelerated GPU MC simulations. We also compare this GPU-ANLM filter with the CPU version and report a threefold to fourfold speedup. The developed GPU-ANLM filter not only can enhance three-dimensional MC photon simulation results but also be a valuable tool for noise reduction in other volumetric images such as MRI and CT scans.

Parallelized Monte Carlo software to efficiently simulate the light propagation in arbitrarily shaped objects and aligned scattering media

A GPU-based Monte Carlo software (MCtet) was developed to calculate the light propagation in arbitrarily shaped objects, like a human tooth, represented by a tetrahedral mesh. A unique feature of MCtet is a concept to realize different kinds of light-sources illuminating the complex-shaped surface of an object, for which no preprocessing step is needed. With this concept, it is also possible to consider photons leaving a turbid media and reentering again in case of a concave object. The correct implementation was shown by comparison with five other Monte Carlo software packages. A hundredfold acceleration compared with central processing units-based programs was found. MCtet can simulate anisotropic light propagation, e.g., by accounting for scattering at cylindrical structures. The important influence of the anisotropic light propagation, caused, e.g., by the tubules in human dentin, is shown for the transmission spectrum through a tooth. It was found that the sensitivity to a change in the oxygen saturation inside the pulp for transmission spectra is much larger if the tubules are considered. Another "light guiding" effect based on a combination of a low scattering and a high refractive index in enamel is described.

New insights into the origin of remote PPG signals in visible light and infrared

Remote photoplethysmography (PPG) is an optical measurement technique with established applications in vital signs monitoring. Recently, the consensual understanding of blood volume variations (BVVs) as the origin of PPG signals was challenged, raising validity concerns about the remote SpO₂ methodology. Recognizing the imperative for new opto-physiological evidence, this investigation supports the volumetric hypothesis with living skin experiments and Monte Carlo simulations of remote PPG-amplitude in visible light (VIS) and infrared (IR). Multilayered models of the skin were developed to simulate the separate contributions from skin layers containing pulsatile arterioles to the PPG signal in the 450–1000 nm range. The simulated spectra were qualitatively compared with observations of the resting and compressed finger pad, and complemented with videocapillaroscopy. Our results indicate that remote PPG systems indeed probe arterial blood. Green wavelengths probe dermal arterioles while red-IR wavelengths also reach subcutaneous BVVs. Owing to stable penetration depths, the red-IR diagnostic window promotes the invariance of SpO₂ measurements to skin non-homogeneities.

Scalable and massively parallel Monte Carlo photon transport simulations for heterogeneous computing platforms

We present a highly scalable Monte Carlo (MC) three-dimensional photon transport simulation platform designed for heterogeneous computing systems. Through the development of a massively parallel MC algorithm using the Open Computing Language framework, this research extends our existing graphics processing unit (GPU)-accelerated MC technique to a highly scalable vendor-independent heterogeneous computing environment, achieving significantly improved performance and software portability. A number of parallel computing techniques are investigated to achieve portable performance over a wide range of computing hardware. Furthermore, multiple thread-level and device-level load-balancing strategies are developed to obtain efficient simulations using multiple central processing units and GPUs.

Algorithm for rapid determination of optical scattering parameters

Preliminary experiments at the NIST Spectral Tri-function Automated Reference Reflectometer (STARR) facility have been conducted with the goal of providing the diffuse optical properties of a solid reference standard with optical properties similar to human skin. Here, we describe an algorithm for determining the best-fit parameters and the statistical uncertainty associated with the measurement. The objective function is determined from the profile log likelihood, including both experimental and Monte Carlo uncertainties. Initially, the log likelihood is determined over a large parameter search box using a relatively small number of Monte Carlo samples such as 2·104. The search area is iteratively reduced to include the 99.9999% confidence region, while doubling the number of samples at each iteration until the experimental uncertainty dominates over the Monte Carlo uncertainty. Typically this occurs by 1.28·106 samples. The log likelihood is then fit to determine a 95% confidence ellipse. The inverse problem requires the values of the log likelihood on many points. Our implementation uses importance sampling to calculate these points on a grid in an efficient manner. Ultimately, the time-to-solution is approximately six times the cost of a Monte Carlo simulation of the radiation transport problem for a single set of parameters with the largest number of photons required. The results are found to be 64 times faster than our implementation of Particle Swarm Optimization.

Massively parallel simulator of optical coherence tomography of inhomogeneous turbid media

BACKGROUND AND OBJECTIVE

An accurate and practical simulator for Optical Coherence Tomography (OCT) could be an important tool to study the underlying physical phenomena in OCT such as multiple light scattering. Recently, many researchers have investigated simulation of OCT of turbid media, e.g., tissue, using Monte Carlo methods. The main drawback of these earlier simulators is the long computational time required to produce accurate results. We developed a massively parallel simulator of OCT of inhomogeneous turbid media that obtains both Class I diffusive reflectivity, due to ballistic and quasi-ballistic scattered photons, and Class II diffusive reflectivity due to multiply scattered photons.

METHODS

This Monte Carlo-based simulator is implemented on graphic processing units (GPUs), using the Compute Unified Device Architecture (CUDA) platform and programming model, to exploit the parallel nature of propagation of photons in tissue. It models an arbitrary shaped sample medium as a tetrahedron-based mesh and uses an advanced importance sampling scheme.

RESULTS

This new simulator speeds up simulations of OCT of inhomogeneous turbid media by about two orders of magnitude. To demonstrate this result, we have compared the computation times of our new parallel simulator and its serial counterpart using two samples of inhomogeneous turbid media. We have shown that our parallel implementation reduced simulation time of OCT of the first sample medium from 407 min to 92 min by using a single GPU card, to 12 min by using 8 GPU cards and to 7 min by using 16 GPU cards. For the second sample medium, the OCT simulation time was reduced from 209 h to 35.6 h by using a single GPU card, and to 4.65 h by using 8 GPU cards, and to only 2 h by using 16 GPU cards. Therefore our new parallel simulator is considerably more practical to use than its central processing unit (CPU)-based counterpart.

CONCLUSIONS

Our new parallel OCT simulator could be a practical tool to study the different physical phenomena underlying OCT, or to design OCT systems with improved performance.

Spatially and spectrally resolved particle swarm optimization for precise optical property estimation using diffuse-reflectance spectroscopy

This paper presents a new approach to estimate optical properties (absorption and scattering coefficients µa and µs) of biological tissues from spatially-resolved spectroscopy measurements. A Particle Swarm Optimization (PSO)-based algorithm was implemented and firstly modified to deal with spatial and spectral resolutions of the data, and to solve the corresponding inverse problem. Secondly, the optimization was improved by fitting exponential decays to the two best points among all clusters of the "particles" randomly distributed all over the parameter space (µs, µa) of possible solutions. The consequent acceleration of all the groups of particles to the "best" curve leads to significant error decrease in the optical property estimation. The study analyzes the estimated optical property error as a function of the various PSO parameter combinations, and several performance criteria such as the cost-function error and the number of iterations in the algorithms proposed. The final one led to error values between ground truth and estimated values of µs and µa less than 6%.

Light distribution modulated diffuse reflectance spectroscopy

Typically, a diffuse reflectance spectroscopy (DRS) system employing a continuous wave light source would need to acquire diffuse reflectances measured at multiple source-detector separations for determining the absorption and reduced scattering coefficients of turbid samples. This results in a multi-fiber probe structure and an indefinite probing depth. Here we present a novel DRS method that can utilize a few diffuse reflectances measured at one source-detector separation for recovering the optical properties of samples. The core of innovation is a liquid crystal (LC) cell whose scattering property can be modulated by the bias voltage. By placing the LC cell between the light source and the sample, the spatial distribution of light in the sample can be varied as the scattering property of the LC cell modulated by the bias voltage, and this would induce intensity variation of the collected diffuse reflectance. From a series of Monte Carlo simulations and phantom measurements, we found that this new light distribution modulated DRS (LDM DRS) system was capable of accurately recover the absorption and scattering coefficients of turbid samples and its probing depth only varied by less than 3% over the full bias voltage variation range. Our results suggest that this LDM DRS platform could be developed to various low-cost, efficient, and compact systems for in-vivo superficial tissue investigation.

Robust near real-time estimation of physiological parameters from megapixel multispectral images with inverse Monte Carlo and random forest regression

PURPOSE

Multispectral imaging can provide reflectance measurements at multiple spectral bands for each image pixel. These measurements can be used for estimation of important physiological parameters, such as oxygenation, which can provide indicators for the success of surgical treatment or the presence of abnormal tissue. The goal of this work was to develop a method to estimate physiological parameters in an accurate and rapid manner suited for modern high-resolution laparoscopic images.

METHODS

While previous methods for oxygenation estimation are based on either simple linear methods or complex model-based approaches exclusively suited for off-line processing, we propose a new approach that combines the high accuracy of model-based approaches with the speed and robustness of modern machine learning methods. Our concept is based on training random forest regressors using reflectance spectra generated with Monte Carlo simulations.

RESULTS

According to extensive in silico and in vivo experiments, the method features higher accuracy and robustness than state-of-the-art online methods and is orders of magnitude faster than other nonlinear regression based methods.

CONCLUSION

Our current implementation allows for near real-time oxygenation estimation from megapixel multispectral images and is thus well suited for online tissue analysis.

Tumor reactive ringlet oxygen approach for Monte Carlo modeling of photodynamic therapy dosimetry

Photodynamic therapy (PDT) is an emergent technique used for the treatment of several diseases. It requires the interaction of three components: a photosensitizer, a light source and tissue oxygen. Knowledge of the biophysical aspects of PDT is important for improving dosimetry protocols and treatment planning. In this paper we propose a model to simulate the spatial and temporal distribution of ground state oxygen ((3)O2), cumulative singlet excited state oxygen ((1)O2)rx and photosensitizer, in this case protoporphyrin IX (PpIX) in an ALA mediated PDT treatment. The results are analyzed in order to improve the treatment dosimetry. We compute the light fluence in the tissue using Monte Carlo simulations running in a GPU system. The concentration of (3)O2, ((1)O2)rx and the photosensitizer are calculated using this light fluence and a set of differential equations describing the photochemical reactions involved in PDT. In the model the initial photosensitizer concentration depends on tissue depth and type, moreover we consider blood vessel damage and its effect in the ground state oxygen concentration in the tissue. We introduce the tumor reactive single oxygen (TRSO) as a new dosimetry metric. It represents the amount of singlet oxygen per tumor volume that reacts, during the treatment, with the molecules in the tumor. This quantity integrates the effect of the light irradiance, the optical properties of the tumor and the normal tissue, the oxygen consumption and supply, and the photosensitizer biodistribution on the skin.

Artificial neural networks for retrieving absorption and reduced scattering spectra from frequency-domain diffuse reflectance spectroscopy at short source-detector separation

Diffuse reflectance spectroscopy (DRS) based on the frequency-domain (FD) technique has been employed to investigate the optical properties of deep tissues such as breast and brain using source to detector separation up to 40 mm. Due to the modeling and system limitations, efficient and precise determination of turbid sample optical properties from the FD diffuse reflectance acquired at a source-detector separation (SDS) of around 1 mm has not been demonstrated. In this study, we revealed that at SDS of 1 mm, acquiring FD diffuse reflectance at multiple frequencies is necessary for alleviating the influence of inevitable measurement uncertainty on the optical property recovery accuracy. Furthermore, we developed artificial neural networks (ANNs) trained by Monte Carlo simulation generated databases that were capable of efficiently determining FD reflectance at multiple frequencies. The ANNs could work in conjunction with a least-square optimization algorithm to rapidly (within 1 second), accurately (within 10%) quantify the sample optical properties from FD reflectance measured at SDS of 1 mm. In addition, we demonstrated that incorporating the steady-state apparatus into the FD DRS system with 1 mm SDS would enable obtaining broadband absorption and reduced scattering spectra of turbid samples in the wavelength range from 650 to 1000 nm.

Non-invasive detection of iron deficiency by fluorescence measurement of erythrocyte zinc protoporphyrin in the lip

Worldwide, more individuals have iron deficiency than any other health problem. Most of those affected are unaware of their lack of iron, in part because detection of iron deficiency has required a blood sample. Here we report a non-invasive method to optically measure an established indicator of iron status, red blood cell zinc protoporphyrin, in the microcirculation of the lower lip. An optical fibre probe is used to illuminate the lip and acquire fluorescence emission spectra in ∼1 min. Dual-wavelength excitation with spectral fitting is used to distinguish the faint zinc protoporphyrin fluorescence from the much greater tissue background fluorescence, providing immediate results. In 56 women, 35 of whom were iron-deficient, the sensitivity and specificity of optical non-invasive detection of iron deficiency were 97% and 90%, respectively. This fluorescence method potentially provides a rapid, easy to use means for point-of-care screening for iron deficiency in resource-limited settings lacking laboratory infrastructure.

Iron deficiency, the most common health problem in the world, has required a blood test for diagnosis. Here, the authors show that iron deficiency can be detected non-invasively and quickly by measuring the fluorescence of red blood cell zinc protoporphyrin in the microcirculation of the lip.

Method for estimating closed-form solutions of the light diffusion equation for turbid media of any boundary shape

This paper reports a method of estimating an approximate closed-form solution to the light diffusion equation for any type of geometry involving Dirichlet's boundary condition with known source location. It is based on estimating the optimum locations of multiple imaginary point sources to cancel the fluence at the extrapolated boundary by constrained optimization using a genetic algorithm. The mathematical derivation of the problem to approach the optimum solution for the direct-current type of diffuse optical systems is described in detail. Our method is first applied to slab geometry and compared with a truncated series solution. After that, it is applied to hemispherical geometry and compared with Monte Carlo simulation results. The method provides a fast and sufficiently accurate fluence distribution for optical reconstruction.

Toward reliable retrieval of functional information of papillary dermis using spatially resolved diffuse reflectance spectroscopy

Spatially resolved diffuse reflectance spectroscopy (SRDRS) has been employed to quantify tissue optical properties and its interrogation volume is majorly controlled by the source-to-detector separations (SDSs). To noninvasively quantify properties of dermis, a SRDRS setup that includes SDS shorter than 1 mm is required. It will be demonstrated in this study that Monte Carlo simulations employing the Henyey-Greenstein phase function cannot always precisely predict experimentally measured diffuse reflectance at such short SDSs, and we speculated this could be caused by the non-negligible backward light scattering at short SDSs that cannot be properly modeled by the Henyey-Greenstein phase function. To accurately recover the optical properties and functional information of dermis using SRDRS, we proposed the use of the modified two-layer (MTL) geometry. Monte Carlo simulations and phantom experiment results revealed that the MTL probing geometry was capable of faithfully recovering the optical properties of upper dermis. The capability of the MTL geometry in probing the upper dermis properties was further verified through a swine study, and it was found that the measurement results were reasonably linked to histological findings. Finally, the MTL probe was utilized to study psoriatic lesions. Our results showed that the MTL probe was sensitive to the physiological condition of tissue volumes within the papillary dermis and could be used in studying the physiology of psoriasis.

Hybrid algorithm for simulating the collimated transmittance of homogeneous stratified turbid media

In this work we describe the development of a program that simulates the propagation of photons through refractive and reflecting optical components such as lenses, mirrors and stops that includes a biological tissue sample as the main issue to be investigated in order to get a simulated value of light distribution, in particular, of the unscattered light. The analysis of the photons that travel through the sample is based on the program Monte Carlo Multi-Layered with some modifications that consider a Gaussian beam as initial source of light. Position, directional cosines and weight of photons exiting the turbid media are used to propagate them through an optical system. As a mean of validation of the program, we selected a typical optical system for measurement of collimated transmittance. Therefore, several tests were carried out to find the optical system that gives the theoretical collimated transmittance at different values of the optical properties of the turbid media. Along this validation, the optimal experimental configuration is found. Using this results, a comparison between the simulated optimal configuration and the experimental set-up was done, by using a colloidal suspension as a turbid media.

Computationally effective solution of the inverse problem in time-of-flight spectroscopy

Photon time-of-flight (PTOF) spectroscopy enables the estimation of absorption and reduced scattering coefficients of turbid media by measuring the propagation time of short light pulses through turbid medium. The present investigation provides a comparison of the assessed absorption and reduced scattering coefficients from PTOF measurements of intralipid 20% and India ink-based optical phantoms covering a wide range of optical properties relevant for biological tissues and dairy products. Three different models are used to obtain the optical properties by fitting to measured temporal profiles: the Liemert-Kienle model (LKM), the diffusion model (DM) and a white Monte-Carlo (WMC) simulation-based algorithm. For the infinite space geometry, a very good agreement is found between the LKM and WMC, while the results obtained by the DM differ, indicating that the LKM can provide accurate estimation of the optical parameters beyond the limits of the diffusion approximation in a computational effective and accurate manner. This result increases the potential range of applications for PTOF spectroscopy within industrial and biomedical applications.

Efficient construction of robust artificial neural networks for accurate determination of superficial sample optical properties

In general, diffuse reflectance spectroscopy (DRS) systems work with photon diffusion models to determine the absorption coefficient μa and reduced scattering coefficient μs' of turbid samples. However, in some DRS measurement scenarios, such as using short source-detector separations to investigate superficial tissues with comparable μa and μs', photon diffusion models might be invalid or might not have analytical solutions. In this study, a systematic workflow of constructing a rapid, accurate photon transport model that is valid at short source-detector separations (SDSs) and at a wide range of sample albedo is revealed. To create such a model, we first employed a GPU (Graphic Processing Unit) based Monte Carlo model to calculate the reflectance at various sample optical property combinations and established a database at high speed. The database was then utilized to train an artificial neural network (ANN) for determining the sample absorption and reduced scattering coefficients from the reflectance measured at several SDSs without applying spectral constraints. The robustness of the produced ANN model was rigorously validated. We evaluated the performance of a successfully trained ANN using tissue simulating phantoms. We also determined the 500-1000 nm absorption and reduced scattering spectra of in-vivo skin using our ANN model and found that the values agree well with those reported in several independent studies.