Rapid and accurate determination of tissue optical properties using least-squares support vector machines

Is Raman the best strategy towards the development of non-invasive continuous glucose monitoring devices for diabetes management?

Diabetes has no well-established cure; thus, its management is critical for avoiding severe health complications involving multiple organs. This requires frequent glycaemia monitoring, and the gold standards for this are fingerstick tests. During the last decades, several blood-withdrawal-free platforms have been being studied to replace this test and to improve significantly the quality of life of people with diabetes (PWD). Devices estimating glycaemia level targeting blood or biofluids such as tears, saliva, breath and sweat, are gaining attention; however, most are not reliable, user-friendly and/or cheap. Given the complexity of the topic and the rise of diabetes, a careful analysis is essential to track scientific and industrial progresses in developing diabetes management systems. Here, we summarize the emerging blood glucose level (BGL) measurement methods and report some examples of devices which have been under development in the last decades, discussing the reasons for them not reaching the market or not being really non-invasive and continuous. After discussing more in depth the history of Raman spectroscopy-based researches and devices for BGL measurements, we will examine if this technique could have the potential for the development of a user-friendly, miniaturized, non-invasive and continuous blood glucose-monitoring device, which can operate reliably, without inter-patient variability, over sustained periods.

Utilizing pulse dynamics for non-invasive Raman spectroscopy of blood analytes

Non-invasive measurement methods offer great benefits in the field of medical diagnostics with molecular-specific techniques such as Raman spectroscopy which is increasingly being used for quantitative measurements of tissue biochemistry in vivo. However, some important challenges still remain for label-free optical spectroscopy to be incorporated into the clinical laboratory for routine testing. In particular, non-analyte-specific variations in tissue properties introduce significant variability of the spectra, thereby preventing reliable calibration. For measurements of blood analytes such as glucose, we propose to decrease the interference from individual tissue characteristics by exploiting the known dynamics of the blood-tissue matrix. We reason that by leveraging the natural blood pulse rhythm, the signals from the blood analytes can be enhanced while those from the static components can be effectively suppressed. Here, time-resolved measurements with subsequent pulse frequency estimation and phase-sensitive detection are proposed to recover the Raman spectra correlated with the dynamic changes at blood-pulse frequency. Pilot in vivo study results are presented to establish the benefits as well as outline the challenges of the proposed method in terms of instrumentation and signal processing.

Machine learning estimation of tissue optical properties

Dynamic, in vivo measurement of the optical properties of biological tissues is still an elusive and critically important problem. Here we develop a technique for inverting a Monte Carlo simulation to extract tissue optical properties from the statistical moments of the spatio-temporal response of the tissue by training a 5-layer fully connected neural network. We demonstrate the accuracy of the method across a very wide parameter space on a single homogeneous layer tissue model and demonstrate that the method is insensitive to parameter selection of the neural network model itself. Finally, we propose an experimental setup capable of measuring the required information in real time in an in vivo environment and demonstrate proof-of-concept level experimental results.

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.

Dual-modality optical biopsy of glioblastomas multiforme with diffuse reflectance and fluorescence: ex vivo retrieval of optical properties

Glioma itself accounts for 80% of all malignant primary brain tumors, and glioblastoma multiforme (GBM) accounts for 55% of such tumors. Diffuse reflectance and fluorescence spectroscopy have the potential to discriminate healthy tissues from abnormal tissues and therefore are promising noninvasive methods for improving the accuracy of brain tissue resection. Optical properties were retrieved using an experimentally evaluated inverse solution. On average, the scattering coefficient is 2.4 times higher in GBM than in low grade glioma (LGG), and the absorption coefficient is 48% higher. In addition, the ratio of fluorescence to diffuse reflectance at the emission peak of 460 nm is 2.6 times higher for LGG while reflectance at 650 nm is 2.7 times higher for GBM. The results reported also show that the combination of diffuse reflectance and fluorescence spectroscopy could achieve sensitivity of 100% and specificity of 90% in discriminating GBM from LGG during ex vivo measurements of 22 sites from seven glioma specimens. Therefore, the current technique might be a promising tool for aiding neurosurgeons in determining the extent of surgical resection of glioma and, thus, improving intraoperative tumor identification for guiding surgical intervention.

Nanoparticle-free tissue-mimicking phantoms with intrinsic scattering

We present an alternative to the conventional approach, phantoms without scattering nanoparticles, where scattering is achieved by the material itself: spherical cavities trapped in a silicone matrix. We describe the properties and fabrication of novel optical phantoms based on a silicone elastomer polydimethylsiloxane (PDMS) and glycerol mixture. Optical properties (absorption coefficient µa , reduced scattering coefficient µs' , and anisotropy factor g) of the fabricated phantoms were retrieved from spectrophotometric measurements (in the 400-1100 nm wavelength range) using the inverse adding-doubling method. The internal structure of the phantoms was studied under a scanning electron microscope, and the chemical composition was assessed by Raman spectroscopy. Composition of the phantom material is reported along with the full characterization of the produced phantoms and ways to control their parameters.

Experimental recovery of intrinsic fluorescence and fluorophore concentration in the presence of hemoglobin: spectral effect of scattering and absorption on fluorescence

The ability to recover the intrinsic fluorescence of biological fluorophores is crucial to accurately identify the fluorophores and quantify their concentrations in the media. Although some studies have successfully retrieved the fluorescence spectral shape of known fluorophores, the techniques usually came with heavy computation costs and did not apply for strongly absorptive media, and the intrinsic fluorescence intensity and fluorophore concentration were not recovered. In this communication, an experimental approach was presented to recover intrinsic fluorescence and concentration of fluorescein in the presence of hemoglobin (Hb). The results indicated that the method was efficient in recovering the intrinsic fluorescence peak and fluorophore concentration with an error of 3% and 10%, respectively. The results also suggested that chromophores with irregular absorption spectra (e.g., Hb) have more profound effects on fluorescence spectral shape than chromophores with monotonic absorption and scattering spectra (e.g., black India ink and polystyrene microspheres).

Spectroscopic approach for dynamic bioanalyte tracking with minimal concentration information

Vibrational spectroscopy has emerged as a promising tool for non-invasive, multiplexed measurement of blood constituents - an outstanding problem in biophotonics. Here, we propose a novel analytical framework that enables spectroscopy-based longitudinal tracking of chemical concentration without necessitating extensive a priori concentration information. The principal idea is to employ a concentration space transformation acquired from the spectral information, where these estimates are used together with the concentration profiles generated from the system kinetic model. Using blood glucose monitoring by Raman spectroscopy as an illustrative example, we demonstrate the efficacy of the proposed approach as compared to conventional calibration methods. Specifically, our approach exhibits a 35% reduction in error over partial least squares regression when applied to a dataset acquired from human subjects undergoing glucose tolerance tests. This method offers a new route at screening gestational diabetes and opens doors for continuous process monitoring without sample perturbation at intermediate time points.

Anatomy of noise in quantitative biological Raman spectroscopy

Raman spectroscopy is a fundamental form of molecular spectroscopy that is widely used to investigate structures and properties of molecules using their vibrational transitions. It relies on inelastic scattering of monochromatic laser light irradiating the specimen. After appropriate filtering the scattered light is dispersed onto a detector to determine the shift from the excitation wavelength, which appears in the form of characteristic spectral patterns. The technique can investigate biological samples and provide real-time diagnosis of diseases. However, despite its intrinsic advantages of specificity and minimal perturbation, the Raman scattered light is typically very weak and limits applications of Raman spectroscopy due to measurement (im)precision, driven by inherent noise in the acquired spectra. In this article, we review the principal noise sources that impact quantitative biological Raman spectroscopy. Further, we discuss how such noise effects can be reduced by innovative changes in the constructed Raman system and appropriate signal processing methods.

Chemical analysis of molecular species through turbid medium

Subsurface analysis of chemical species is imperative for biomedical diagnostics and imaging, homeland security, and pharmaceutical and other industries; however, the access to the object of interest is often obscured by an optically scattering medium which limits the ability to inspect the chemical composition of the sample. In this report, we employ coherent Raman microspectroscopy in a combination with a hierarchical cluster analysis to mitigate the effect of scattering and demonstrate the identification of multiple chemical species.

Application of Raman spectroscopy to identify microcalcifications and underlying breast lesions at stereotactic core needle biopsy

Microcalcifications are a feature of diagnostic significance on a mammogram and a target for stereotactic breast needle biopsy. Here, we report development of a Raman spectroscopy technique to simultaneously identify microcalcification status and diagnose the underlying breast lesion, in real-time, during stereotactic core needle biopsy procedures. Raman spectra were obtained ex vivo from 146 tissue sites from fresh stereotactic breast needle biopsy tissue cores from 33 patients, including 50 normal tissue sites, 77 lesions with microcalcifications, and 19 lesions without microcalcifications, using a compact clinical system. The Raman spectra were modeled on the basis of the breast tissue components, and a support vector machine framework was used to develop a single-step diagnostic algorithm to distinguish normal tissue, fibrocystic change (FCC), fibroadenoma, and breast cancer, in the absence and presence of microcalcifications. This algorithm was subjected to leave-one-site-out cross-validation, yielding a positive predictive value, negative predictive value, sensitivity, and specificity of 100%, 95.6%, 62.5%, and 100% for diagnosis of breast cancer (with or without microcalcifications) and an overall accuracy of 82.2% for classification into specific categories of normal tissue, FCC, fibroadenoma, or breast cancer (with and without microcalcifications). Notably, the majority of breast cancers diagnosed are ductal carcinoma in situ (DCIS), the most common lesion associated with microcalcifications, which could not be diagnosed using previous Raman algorithm(s). Our study shows the potential of Raman spectroscopy to concomitantly detect microcalcifications and diagnose associated lesions, including DCIS, and thus provide real-time feedback to radiologists during such biopsy procedures, reducing nondiagnostic and false-negative biopsies.

A hybrid method to recognize 3D object

A hybrid method using the support vector machine (SVM) correlation filter and the phase-shift interferometry (PSI) holography is proposed to recognize 3D object, which can improve the correct decision rate and resist the distortion of object rotation and noise. The different images of two types of both in-plane and out-of-plane rotated object recorded by digital holography are reconstructed. The reconstructed images of two types are selected to synthesize the SVM correlation filter, respectively. To compare the correct decision rates of the SVM correlation filter with other three ones, it is found that the experimental result is better in rotation resistance and noise tolerance.

Investigation of noise-induced instabilities in quantitative biological spectroscopy and its implications for noninvasive glucose monitoring

Over the past decade, optical spectroscopy has been employed in combination with multivariate chemometric models to investigate a wide variety of diseases and pathological conditions, primarily due to its excellent chemical specificity and lack of sample preparation requirements. Despite promising results in several proof-of-concept studies, its translation to the clinical setting has often been hindered by inadequate accuracy of the conventional spectroscopic models. To address this issue and the possibility of curved (nonlinear) effects in the relationship between the concentrations of the analyte of interest and the mixture spectra (due to fluctuations in sample and environmental conditions), support vector machine-based least-squares nonlinear regression (LS-SVR) has been recently proposed. In this paper, we investigate the robustness of this methodology to noise-induced instabilities and present an analytical formula for estimating modeling precision as a function of measurement noise and model parameters. This formalism can be readily used to evaluate uncertainty in information extracted from spectroscopic measurements, particularly important for rapid-acquisition biomedical applications. Subsequently, using field data (Raman spectra) acquired from a glucose clamping study on an animal model subject, we perform the first systematic investigation of the relative effect of additive interference components (namely, noise in prediction spectra, calibration spectra, and calibration concentrations) on the prediction error of nonlinear spectroscopic models. Our results show that the LS-SVR method gives more accurate results and is substantially more robust to additive noise when compared with conventional regression methods such as partial least-squares regression (PLS), when careful selection of the LS-SVR model parameters are performed. We anticipate that these results will be useful for uncertainty estimation in similar biomedical applications where the precision of measurements and its response to noise in the data set is as important, if not more so, than the generic accuracy level.

Joint derivation method for determining optical properties based on steady-state spatially resolved diffuse reflectance measurement at small source-detector separations and large reduced albedo range: theory and simulation

Accurate determination of the optical properties (the absorption coefficient μ(a) and the reduced scattering coefficient μ(s) (')) of tissues is very important in a variety of diagnostic and therapeutic procedures. Optical diffusion theory is frequently used as the forward model for describing the photon transfer in media with large reduced albedos (a(')) and in large source-detector separations (SDS). Several other methods (PN approximation, hybrid diffusion-P3 approximation) have also been published that describe photon transfer in media with low a(') or small SDSs. We studied the theoretical models for the steady-state spatially resolved diffuse reflectance measurement to accurately determine μ(a) and μ(s) (') at large a(') range but small SDSs. Instead of using a single model, a joint derivation method is proposed. The developed method uses one of the best aforementioned theoretical methods separately in five ranges of a(') determined from several forward models. In the region of small SDSs (the range between 0.4 and 8 mm) and large a(') range (between 0.5 and 0.99), the best theoretical derivation model was determined. The results indicate that the joint derivation method can improve the derivation accuracy and that a(') range can be determined by the steady-state spatially resolved diffuse reflectance measurement.

Incorporation of support vector machines in the LIBS toolbox for sensitive and robust classification amidst unexpected sample and system variability

Despite the intrinsic elemental analysis capability and lack of sample preparation requirements, laser-induced breakdown spectroscopy (LIBS) has not been extensively used for real-world applications, e.g., quality assurance and process monitoring. Specifically, variability in sample, system, and experimental parameters in LIBS studies present a substantive hurdle for robust classification, even when standard multivariate chemometric techniques are used for analysis. Considering pharmaceutical sample investigation as an example, we propose the use of support vector machines (SVM) as a nonlinear classification method over conventional linear techniques such as soft independent modeling of class analogy (SIMCA) and partial least-squares discriminant analysis (PLS-DA) for discrimination based on LIBS measurements. Using over-the-counter pharmaceutical samples, we demonstrate that the application of SVM enables statistically significant improvements in prospective classification accuracy (sensitivity), because of its ability to address variability in LIBS sample ablation and plasma self-absorption behavior. Furthermore, our results reveal that SVM provides nearly 10% improvement in correct allocation rate and a concomitant reduction in misclassification rates of 75% (cf. PLS-DA) and 80% (cf. SIMCA)-when measurements from samples not included in the training set are incorporated in the test data-highlighting its robustness. While further studies on a wider matrix of sample types performed using different LIBS systems is needed to fully characterize the capability of SVM to provide superior predictions, we anticipate that the improved sensitivity and robustness observed here will facilitate application of the proposed LIBS-SVM toolbox for screening drugs and detecting counterfeit samples, as well as in related areas of forensic and biological sample analysis.

Portable optical fiber probe-based spectroscopic scanner for rapid cancer diagnosis: a new tool for intraoperative margin assessment

There continues to be a significant clinical need for rapid and reliable intraoperative margin assessment during cancer surgery. Here we describe a portable, quantitative, optical fiber probe-based, spectroscopic tissue scanner designed for intraoperative diagnostic imaging of surgical margins, which we tested in a proof of concept study in human tissue for breast cancer diagnosis. The tissue scanner combines both diffuse reflectance spectroscopy (DRS) and intrinsic fluorescence spectroscopy (IFS), and has hyperspectral imaging capability, acquiring full DRS and IFS spectra for each scanned image pixel. Modeling of the DRS and IFS spectra yields quantitative parameters that reflect the metabolic, biochemical and morphological state of tissue, which are translated into disease diagnosis. The tissue scanner has high spatial resolution (0.25 mm) over a wide field of view (10 cm × 10 cm), and both high spectral resolution (2 nm) and high spectral contrast, readily distinguishing tissues with widely varying optical properties (bone, skeletal muscle, fat and connective tissue). Tissue-simulating phantom experiments confirm that the tissue scanner can quantitatively measure spectral parameters, such as hemoglobin concentration, in a physiologically relevant range with a high degree of accuracy (<5% error). Finally, studies using human breast tissues showed that the tissue scanner can detect small foci of breast cancer in a background of normal breast tissue. This tissue scanner is simpler in design, images a larger field of view at higher resolution and provides a more physically meaningful tissue diagnosis than other spectroscopic imaging systems currently reported in literatures. We believe this spectroscopic tissue scanner can provide real-time, comprehensive diagnostic imaging of surgical margins in excised tissues, overcoming the sampling limitation in current histopathology margin assessment. As such it is a significant step in the development of a platform technology for intraoperative management of cancer, a clinical problem that has been inadequately addressed to date.

Hybrid diffusion-P3 equation in N-layered turbid media: steady-state domain

This paper discusses light propagation in N-layered turbid media. The hybrid diffusion-P3 equation is solved for an N-layered finite or infinite turbid medium in the steady-state domain for one point source using the extrapolated boundary condition. The Fourier transform formalism is applied to derive the analytical solutions of the fluence rate in Fourier space. Two inverse Fourier transform methods are developed to calculate the fluence rate in real space. In addition, the solutions of the hybrid diffusion-P3 equation are compared to the solutions of the diffusion equation and the Monte Carlo simulation. For the case of small absorption coefficients, the solutions of the N-layered diffusion equation and hybrid diffusion-P3 equation are almost equivalent and are in agreement with the Monte Carlo simulation. For the case of large absorption coefficients, the model of the hybrid diffusion-P3 equation is more precise than that of the diffusion equation. In conclusion, the model of the hybrid diffusion-P3 equation can replace the diffusion equation for modeling light propagation in the N-layered turbid media for a wide range of absorption coefficients.