Texture analysis and classification in coherent anti-Stokes Raman scattering (CARS) microscopy images for automated detection of skin cancer

Two-photon microendoscope for label-free imaging in stereotactic neurosurgery

We demonstrate a gradient refractive index (GRIN) microendoscope with an outer diameter of ∼1.2 mm and a length of ∼186 mm that can fit into a stereotactic surgical cannula. Two photon imaging at an excitation wavelength of 900 nm showed a field of view of ∼180 microns and a lateral and axial resolution of 0.86 microns and 9.6 microns respectively. The microendoscope was tested by imaging autofluorescence and second harmonic generation (SHG) in label-free human brain tissue. Furthermore, preliminary image analysis indicates that image classification models can predict if an image is from the subthalamic nucleus or the surrounding tissue using conventional, bench-top two-photon autofluorescence.

Label-free multiphoton microscopy enables histopathological assessment of colorectal liver metastases and supports automated classification of neoplastic tissue

As the state of resection margins is an important prognostic factor after extirpation of colorectal liver metastases, surgeons aim to obtain negative margins, sometimes elaborated by resections of the positive resection plane after intraoperative frozen sections. However, this is time consuming and results sometimes remain unclear during surgery. Label-free multimodal multiphoton microscopy (MPM) is an optical technique that retrieves morpho-chemical information avoiding all staining and that can potentially be performed in real-time. Here, we investigated colorectal liver metastases and hepatic tissue using a combination of three endogenous nonlinear signals, namely: coherent anti-Stokes Raman scattering (CARS) to visualize lipids, two-photon excited fluorescence (TPEF) to visualize cellular patterns, and second harmonic generation (SHG) to visualize collagen fibers. We acquired and analyzed over forty thousand MPM images of metastatic and normal liver tissue of 106 patients. The morphological information with biochemical specificity produced by MPM allowed discriminating normal liver from metastatic tissue and discerning the tumor borders on cryosections as well as formalin-fixed bulk tissue. Furthermore, automated tissue type classification with a correct rate close to 95% was possible using a simple approach based on discriminant analysis of texture parameters. Therefore, MPM has the potential to increase the precision of resection margins in hepatic surgery of metastases without prolonging surgical intervention.

Development of a SERS based cancer diagnosis approach employing cryosectioned thyroid tissue samples on PDMS

An efficient SERS based novel analytical approach named Cryosectioned-PDMS was developed systematically and evaluated applying on 64 thyroid biopsy samples. To utilize thyroid biopsy samples, a 20-μl volume of h-AgNPs suspension was dropped on a 5-μm thick cryosectioned biopsy specimen placed on the PDMS coated glass slide. The SERS spectra from a 10 × 10 points array acquired by mapping 22.5 μm × 22.5 μm sized area from suspended dried droplets placed on the tissue surface. The probability of correctly predicted performance for diagnosis of malignant, benign and healthy tissues was resulted in the accuracy of 100 % for the spectral bands at 667, 724, 920, 960, 1052, 1096, 1315 and 1457 cm^-1 using PCA-fed LDA machine learning. The Cryosectioned-PDMS biophotonic approach with PCA-LDA predictive model demonstrated that the vibrational signatures can accurately recognize the fingerprint of cancer pathology from a healthy one with a simple and fast sample preparation methodology.

Types of spectroscopy and microscopy techniques for cancer diagnosis: a review

Cancer is a life-threatening disease that has claimed the lives of many people worldwide. With the current diagnostic methods, it is hard to determine cancer at an early stage, due to its versatile nature and lack of genomic biomarkers. The rapid development of biophotonics has emerged as a potential tool in cancer detection and diagnosis. Using the fluorescence, scattering, and absorption characteristics of cells and tissues, it is possible to detect cancer at an early stage. The diagnostic techniques addressed in this review are highly sensitive to the chemical and morphological changes in the cell and tissue during disease progression. These changes alter the fluorescence signal of the cell/tissue and are detected using spectroscopy and microscopy techniques including confocal and two-photon fluorescence (TPF). Further, second harmonic generation (SHG) microscopy reveals the morphological changes that occurred in non-centrosymmetric structures in the tissue, such as collagen. Again, Raman spectroscopy is a non-destructive method that provides a fingerprinting technique to differentiate benign and malignant tissue based on Raman signal. Photoacoustic microscopy and spectroscopy of tissue allow molecule-specific detection with high spatial resolution and penetration depth. In addition, terahertz spectroscopic studies reveal the variation of tissue water content during disease progression. In this review, we address the applications of spectroscopic and microscopic techniques for cancer detection based on the optical properties of the tissue. The discussed state-of-the-art techniques successfully determines malignancy to its rapid diagnosis.

Imaging approaches for monitoring three-dimensional cell and tissue culture systems

The past decade has seen an increasing demand for more complex, reproducible and physiologically relevant tissue cultures that can mimic the structural and biological features of living tissues. Monitoring the viability, development and responses of such tissues in real-time are challenging due to the complexities of cell culture physical characteristics and the environments in which these cultures need to be maintained in. Significant developments in optics, such as optical manipulation, improved detection and data analysis, have made optical imaging a preferred choice for many three-dimensional (3D) cell culture monitoring applications. The aim of this review is to discuss the challenges associated with imaging and monitoring 3D tissues and cell culture, and highlight topical label-free imaging tools that enable bioengineers and biophysicists to non-invasively characterise engineered living tissues.

Image-to-image translation of label-free molecular vibrational images for a histopathological review using the UNet+/seg-cGAN model

Translating images generated by label-free microscopy imaging, such as Coherent Anti-Stokes Raman Scattering (CARS), into more familiar clinical presentations of histopathological images will help the adoption of real-time, spectrally resolved label-free imaging in clinical diagnosis. Generative adversarial networks (GAN) have made great progress in image generation and translation, but have been criticized for lacking precision. In particular, GAN has often misinterpreted image information and identified incorrect content categories during image translation of microscopy scans. To alleviate this problem, we developed a new Pix2pix GAN model that simultaneously learns classifying contents in the images from a segmentation dataset during the image translation training. Our model integrates UNet+ with seg-cGAN, conditional generative adversarial networks with partial regularization of segmentation. Technical innovations of the UNet+/seg-cGAN model include: (1) replacing UNet with UNet+ as the Pix2pix cGAN's generator to enhance pattern extraction and richness of the gradient, and (2) applying the partial regularization strategy to train a part of the generator network as the segmentation sub-model on a separate segmentation dataset, thus enabling the model to identify correct content categories during image translation. The quality of histopathological-like images generated based on label-free CARS images has been improved significantly.

From Raman to SESORRS: moving deeper into cancer detection and treatment monitoring

Raman spectroscopy is a non-invasive technique that allows specific chemical information to be obtained from various types of sample. The detailed molecular information that is present in Raman spectra permits monitoring of biochemical changes that occur in diseases, such as cancer, and can be used for the early detection and diagnosis of the disease, for monitoring treatment, and to distinguish between cancerous and non-cancerous biological samples. Several techniques have been developed to enhance the capabilities of Raman spectroscopy by improving detection sensitivity, reducing imaging times and increasing the potential applicability for in vivo analysis. The different Raman techniques each have their own advantages that can accommodate the alternative detection formats, allowing the techniques to be applied in several ways for the detection and diagnosis of cancer. This feature article discusses the various forms of Raman spectroscopy, how they have been applied for cancer detection, and the adaptation of the techniques towards their use for in vivo cancer detection and in clinical diagnostics. Despite the advances in Raman spectroscopy, the clinical application of the technique is still limited and certain challenges must be overcome to enable clinical translation. We provide an outlook on the future of the techniques in this area and what we believe is required to allow the potential of Raman spectroscopy to be achieved for clinical cancer diagnostics.

Artificial intelligence-augmented, label-free molecular imaging method for tissue identification, cancer diagnosis, and cancer margin detection

Label-free high-resolution molecular and cellular imaging strategies for intraoperative use are much needed, but not yet available. To fill this void, we developed an artificial intelligence-augmented molecular vibrational imaging method that integrates label-free and subcellular-resolution coherent anti-stokes Raman scattering (CARS) imaging with real-time quantitative image analysis via deep learning (artificial intelligence-augmented CARS or iCARS). The aim of this study was to evaluate the capability of the iCARS system to identify and differentiate the parathyroid gland and recurrent laryngeal nerve (RLN) from surrounding tissues and detect cancer margins. This goal was successfully met.

Coherent Raman scattering microscopy for chemical imaging of biological systems

Coherent Raman scattering (CRS) processes, including both the coherent anti-Stokes Raman scattering and stimulated Raman scattering, have been utilized in state-of-the-art microscopy platforms for chemical imaging of biological samples. The key advantage of CRS microscopy over fluorescence microscopy is label-free, which is an attractive characteristic for modern biological and medical sciences. Besides, CRS has other advantages such as higher selectivity to metabolites, no photobleaching, and narrow peak width. These features have brought fast-growing attention to CRS microscopy in biological research. In this review article, we will first briefly introduce the history of CRS microscopy, and then explain the theoretical background of the CRS processes in detail using the classical approach. Next, we will cover major instrumentation techniques of CRS microscopy. Finally, we will enumerate examples of recent applications of CRS imaging in biological and medical sciences.

Characterization of Mueller matrix elements for classifying human skin cancer utilizing random forest algorithm

SIGNIFICANCE

The Mueller matrix decomposition method is widely used for the analysis of biological samples. However, its presumed sequential appearance of the basic optical effects (e.g., dichroism, retardance, and depolarization) limits its accuracy and application.

AIM

An approach is proposed for detecting and classifying human melanoma and non-melanoma skin cancer lesions based on the characteristics of the Mueller matrix elements and a random forest (RF) algorithm.

APPROACH

In the proposal technique, 669 data points corresponding to the 16 elements of the Mueller matrices obtained from 32 tissue samples with squamous cell carcinoma (SCC), basal cell carcinoma (BCC), melanoma, and normal features are input into an RF classifier as predictors.

RESULTS

The results show that the proposed model yields an average precision of 93%. Furthermore, the classification results show that for biological tissues, the circular polarization properties (i.e., elements m44, m34, m24, and m14 of the Mueller matrix) dominate the linear polarization properties (i.e., elements m13, m31, m22, and m41 of the Mueller matrix) in determining the classification outcome of the trained classifier.

CONCLUSIONS

Overall, our study provides a simple, accurate, and cost-effective solution for developing a technique for classification and diagnosis of human skin cancer.

Label-free multiphoton imaging allows brain tumor recognition based on texture analysis-a study of 382 tumor patients

Background

Label-free multiphoton microscopy has been suggested for intraoperative recognition and delineation of brain tumors. For any future clinical application, appropriate approaches for image acquisition and analysis have to be developed. Moreover, an evaluation of the reliability of the approach, taking into account inter- and intrapatient variability, is needed.

Methods

Coherent anti-Stokes Raman scattering (CARS), two-photon excited fluorescence (TPEF), and second-harmonic generation were acquired on cryosections of brain tumors of 382 patients and 28 human nontumor brain samples. Texture parameters of those images were calculated and used as input for linear discriminant analysis.

Results

The combined analysis of texture parameters of the CARS and TPEF signal proved to be most suited for the discrimination of nontumor brain versus brain tumors (low- and high-grade astrocytoma, oligodendroglioma, glioblastoma, recurrent glioblastoma, brain metastases of lung, colon, renal, and breast cancer and of malignant melanoma) leading to a correct rate of 96% (sensitivity: 96%, specificity: 100%). To approximate the clinical setting, the results were validated on 42 fresh, unfixed tumor biopsies. 82% of the tumors and, most important, all of the nontumor samples were correctly recognized. An image resolution of 1 µm was sufficient to distinguish brain tumors and nontumor brain. Moreover, the vast majority of single fields of view of each patient’s sample were correctly classified with high probabilities, which is important for clinical translation.

Conclusion

Label-free multiphoton imaging might allow fast and accurate intraoperative delineation of primary and secondary brain tumors in combination with endoscopic systems.

Texture classification of MR images of the brain in ALS using M-CoHOG: A multi-center study

Gradient-based texture analysis methods have become popular in computer vision and image processing and has many applications including medical image analysis. This motivates us to develop a texture feature extraction method to discriminate Amyotrophic Lateral Sclerosis (ALS) patients from controls. But, the lack of data in ALS research is a major constraint and can be mitigated by using data from multiple centers. However, multi-center data gives some other challenges such as differing scanner parameters and variation in intensity of the medical images, which motivate the development of the proposed method. To investigate these challenges, we propose a gradient-based texture feature extraction method called Modified Co-occurrence Histograms of Oriented Gradients (M-CoHOG) to extract texture features from 2D Magnetic Resonance Images (MRI). We also propose a new feature-normalization technique before feeding the normalized M-CoHOG features into an ensemble of classifiers, which can accommodate for variation of data from different centers. ALS datasets from four different centers are used in the experiments. We analyze the classification accuracy of single center data as well as that arising from multiple centers. It is observed that the extracted texture features from downsampled images are more significant in distinguishing between patients and controls. Moreover, using an ensemble of classifiers shows improvement in classification accuracy over a single classifier in multi-center data. The proposed method outperforms the state-of-the-art methods by a significant margin.

Portable all-fiber dual-output widely tunable light source for coherent Raman imaging

We present a rapidly tunable dual-output all-fiber light source for coherent Raman imaging, based on a dispersively matched mode-locked laser pumping a parametric oscillator. Output pump and Stokes pulses with a maximal power of 170 and 400 mW, respectively, and equal durations of 7 ps could be generated. The tuning mechanism required no mechanical delay line, enabling all-electronic arbitrary wavelength switching across more than 2700  cm - 1 in less than 5 ms. The compact setup showed a reliable operation despite mechanical shocks of more than 25  m / s 2 and is, thus, well suited for operation in a mobile cart. Imaging mouse and human skin tissue with both the portable light source and a commercial laboratory-bound reference system yielded qualitatively equal results and verified the portable light source being well suited for coherent Raman microscopy.

Whole-Section Tumor Micro-Architecture Analysis by a Two-Dimensional Phasor-Based Approach Applied to Polarization-Dependent Second Harmonic Imaging

Second Harmonic Generation (SHG) microscopy has gained much interest in the histopathology field since it allows label-free imaging of tissues simultaneously providing information on their morphology and on the collagen microarchitecture, thereby highlighting the onset of pathologies and diseases. A wide request of image analysis tools is growing, with the aim to increase the reliability of the analysis of the huge amount of acquired data and to assist pathologists in a user-independent way during their diagnosis. In this light, we exploit here a set of phasor-parameters that, coupled to a 2-dimensional phasor-based approach (μMAPPS, Microscopic Multiparametric Analysis by Phasor projection of Polarization-dependent SHG signal) and a clustering algorithm, allow to automatically recover different collagen microarchitectures in the tissues extracellular matrix. The collagen fibrils microscopic parameters (orientation and anisotropy) are analyzed at a mesoscopic level by quantifying their local spatial heterogeneity in histopathology sections (few mm in size) from two cancer xenografts in mice, in order to maximally discriminate different collagen organizations, allowing in this case to identify the tumor area with respect to the surrounding skin tissue. We show that the "fibril entropy" parameter, which describes the tissue order on a selected spatial scale, is the most effective in enlightening the tumor edges, opening the possibility of their automatic segmentation. Our method, therefore, combined with tissue morphology information, has the potential to become a support to standard histopathology in diseases diagnosis.

Hierarchical deep convolutional neural networks combine spectral and spatial information for highly accurate Raman-microscopy-based cytopathology

Hierarchical variants of so-called deep convolutional neural networks (DCNNs) have facilitated breakthrough results for numerous pattern recognition tasks in recent years. We assess the potential of these novel whole-image classifiers for Raman-microscopy-based cytopathology. Conceptually, DCNNs facilitate a flexible combination of spectral and spatial information for classifying cellular images as healthy or cancer-affected cells. As we demonstrate, this conceptual advantage translates into practice, where DCNNs exceed the accuracy of both conventional classifiers based on pixel spectra as well as classifiers based on morphological features extracted from Raman microscopic images. Remarkably, accuracies exceeding those of all previously proposed classifiers are obtained while using only a small fraction of the spectral information provided by the dataset. Overall, our results indicate a high potential for DCNNs in medical applications of not just Raman, but also infrared microscopy.

Accuracy of Raman spectroscopy for differentiating skin cancer from normal tissue

BACKGROUND

Raman spectroscopy could be applied to distinguish tumor from normal tissues. This meta-analysis assessed the accuracy of Raman spectroscopy in differentiating skin cancer from normal tissue.

METHODS

PubMed, Embase, Cochrane Library, and CNKI were searched to identify suitable studies before Februray 4th, 2018. We estimated the pooled sensitivity, specificity, positive, and negative likelihood ratios, diagnostic odds ratio, and constructed summary receiver-operating characteristics curves to identify the accuracy of Raman spectroscopy in differentiating skin cancer from normal tissue.

RESULTS

A total of 12 studies with 2461 spectra were included. For basal cell skin cancer (BCC) ex vivo detection, the pooled sensitivity and specificity were 0.99 (95% confidence interval [CI] 0.97-0.99) and 0.96 (95% CI 0.95-0.97), respectively. The area under the curve (AUC) was 0.9837. For BCC in vivo detection, the pooled sensitivity and specificity were 0.69 (95% CI 0.61-0.76) and 0.85 (95% CI 0.82-0.87), respectively. The AUC was 0.9213. For melanoma (MM) ex vivo detection, the pooled sensitivity and specificity were 1.00 (95% CI 0.91-1.00) and 0.98 (95% CI 0.95-1.00), respectively. The AUC was 0.9914. For MM in vivo detection, the sensitivity (0.93) and the specificity (0.96) balanced relatively well. For squamous cell skin cancer (SCC) ex vivo detection, the pooled sensitivity and specificity were 0.96 (95% CI 0.81-1.00) and 1.00 (95% CI 0.92-1.00), respectively. For SCC in vivo detection, the sensitivity was 0.81 (95% CI 0.70-0.90) and the specificity was 0.89 (95% CI 0.86-0.91).

CONCLUSION

This meta-analysis suggested that Raman spectroscopy could be an effective and accurate tool for differentiating BCC, MM, SCC from normal tissue, which would assist us in the diagnosis and treatment of skin cancer.

Label-Free Molecular Imaging of Biological Cells and Tissues by Linear and Nonlinear Raman Spectroscopic Approaches

Raman spectroscopy is an emerging technique in bioanalysis and imaging of biomaterials owing to its unique capability of generating spectroscopic fingerprints. Imaging cells and tissues by Raman microspectroscopy represents a nondestructive and label-free approach. All components of cells or tissues contribute to the Raman signals, giving rise to complex spectral signatures. Resonance Raman scattering and surface-enhanced Raman scattering can be used to enhance the signals and reduce the spectral complexity. Raman-active labels can be introduced to increase specificity and multimodality. In addition, nonlinear coherent Raman scattering methods offer higher sensitivities, which enable the rapid imaging of larger sampling areas. Finally, fiber-based imaging techniques pave the way towards in vivo applications of Raman spectroscopy. This Review summarizes the basic principles behind medical Raman imaging and its progress since 2012.

Implementation of fluorescence confocal mosaicking microscopy by "early adopter" Mohs surgeons and dermatologists: recent progress

Confocal mosaicking microscopy (CMM) enables rapid imaging of large areas of fresh tissue ex vivo without the processing that is necessary for conventional histology. When performed in fluorescence mode using acridine orange (nuclear specific dye), it enhances nuclei-to-dermis contrast that enables detection of all types of basal cell carcinomas (BCCs), including micronodular and thin strands of infiltrative types. So far, this technique has been mostly validated in research settings for the detection of residual BCC tumor margins with high sensitivity of 89% to 96% and specificity of 99% to 89%. Recently, CMM has advanced to implementation and testing in clinical settings by “early adopter” Mohs surgeons, as an adjunct to frozen section during Mohs surgery. We summarize the development of CMM guided imaging of ex vivo skin tissues from bench to bedside. We also present its current state of application in routine clinical workflow not only for the assessment of residual BCC margins in the Mohs surgical setting but also for some melanocytic lesions and other skin conditions in clinical dermatology settings. Last, we also discuss the potential limitations of this technology as well as future developments. As this technology advances further, it may serve as an adjunct to standard histology and enable rapid surgical pathology of skin cancers at the bedside.

Differentiating responses of lung cancer cell lines to Doxorubicin exposure: in vitro Raman micro spectroscopy, oxidative stress and bcl-2 protein expression

The potential of Raman micro spectroscopy as an in vitro, non-invasive tool for clinical applications has been demonstrated in recent years, specifically for cancer research. To further illustrate its potential as a high content and label free technique, it is important to show its capability to elucidate drug mechanisms of action and cellular resistances. In this study, cytotoxicity assays were employed to establish the toxicity profiles for 24 hr exposure of lung cancer cell lines, A549 and Calu-1, to the commercially available drug, doxorubicin (DOX). Raman spectroscopy, coupled with Confocal Laser Scanning Microscopy and Flow Cytometry, was used to track the DOX mechanism of action, at a subcellular level, and to study the mechanisms of cellular resistance to DOX. Biomarkers related to the drug mechanism of action and cellular resistance to apoptosis, namely reactive oxygen species (ROS) and bcl-2 protein expression, respectively, were also measured and correlated to Raman spectral profiles. Calu-1 cells are shown to exhibit spectroscopic signatures of both direct DNA damage due to intercalation in the nucleus and indirect damage due to oxidative stress in the cytoplasm, whereas the A549 cell line only exhibits signatures of the former mechanism of action. PCA of nucleolar, nuclear and cytoplasmic regions of A549 and Calu-1 with corresponding loadings of PC1 and PC2.