Abstract In this study, optical photothermal infrared (O-PTIR) spectroscopy combined with machine learning algorithms were used to evaluate 46 tissue cores of surgically resected cervical lymph nodes, some of which harboured oral squamous cell carcinoma nodal metastasis. The ratios obtained between O-PTIR chemical images at 1252 cm−1 and 1285 cm−1 were able to reveal morphological details from tissue samples that are comparable to the information achieved by a pathologist’s interpretation of optical microscopy of haematoxylin and eosin (H&E) stained samples. Additionally, when used as input data for a hybrid convolutional neural network (CNN) and random forest (RF) analyses, these yielded sensitivities, specificities and precision of 98.6 ± 0.3%, 92 ± 4% and 94 ± 5%, respectively, and an area under receiver operator characteristic (AUC) of 94 ± 2%. Our findings show the potential of O-PTIR technology as a tool to study cancer on tissue samples.

BackgroundEstimation of prognosis of oral squamous cell carcinoma (OSCC) is inaccurate prior to surgery, only being effected following subsequent pathological analysis of the primary tumour and excised lymph nodes. Consequently, a proportion of patients are overtreated, with an increase in morbidity, or undertreated, with inadequate margins and risk of recurrence. We hypothesise that it is possible to accurately characterise clinical outcomes from infrared spectra arising from diagnostic biopsies. In this first step, we correlate survival with IR spectra derived from the primary tumour.MethodsInfrared spectra were collected from tumour tissue from 29 patients with OSCC and subject to classification modelling.ResultsThe model had a median AUROC of 0.89 with regard to prognosis, a median specificity of 0.83, and a hazard ratio of 6.29 in univariate Cox proportional hazard modelling.ConclusionThe data suggest that FTIR spectra may be a useful early biomarker of prognosis in OSCC. Estimation of prognosis of oral squamous cell carcinoma (OSCC) is inaccurate prior to surgery leading to over treatment in a proportion of patients. Infrared spectra were collected from OSCC tumour tissue and subject to classification modelling. The model had a median AUROC of 0.89 with regard to prognosis, a median specificity of 0.83 and a hazard ratio of 6.29 in univariate Cox proportional hazard modelling. The data suggest that FTIR spectra may be a useful early biomarker of prognosis in OSCC.image

A machine learning algorithm (MLA) has predicted the prognosis of oral potentially malignant lesions and discriminated between lymph node tissue and metastatic oral squamous cell carcinoma (OSCC). The MLA analyses metrics, which are ratios of Fourier transform infrared absorbances, and identifies key wavenumbers that can be associated with molecular biomarkers. The wider efficacy of the MLA is now shown in the more complex primary OSCC tumour setting, where it is able to identify seven types of tissue. Three epithelial and four non-epithelial tissue types were discriminated from each other with sensitivities between 82% and 96% and specificities between 90% and 99%. The wavenumbers involved in the five best discriminating metrics for each tissue type were tightly grouped, indicating that small changes in the spectral profiles of the different tissue types are important. The number of samples used in this study was small, but the information will provide a basis for further, larger investigations.

A regression-based fusion algorithm has been used to merge hyperspectral Fourier transform infrared (FTIR) data with an H & E image of oral squamous cell carcinoma metastases in cervical lymphoid nodal tissue. This provides insight into the success of the ratio of FTIR absorbances at 1252 cm(-1) and 1285 cm(-1) in discriminating between these tissue types. The success is due to absorbances at these two wavenumbers being dominated by contributions from DNA and collagen, respectively. A pixel-by-pixel fit of the fused spectra to the FTIR spectra of collagen, DNA and cytokeratin reveals the contributions of these molecules to the tissue at high spatial resolution.

A machine learning algorithm (MLA) has been applied to a Fourier transform infrared spectroscopy (FTIR) dataset previously analysed with a principal component analysis (PCA) linear discriminant analysis (LDA) model. This comparison has confirmed the robustness of FTIR as a prognostic tool for oral epithelial dysplasia (OED). The MLA is able to predict malignancy with a sensitivity of 84 ± 3% and a specificity of 79 ± 3%. It provides key wavenumbers that will be important for the development of devices that can be used for improved prognosis of OED.

It is shown that a pixel-level image fusion technique can produce images that combine the spatial resolution of optical microscopy images of haematoxylin and eosin (H&E) stained tissue with the chemical information in Fourier transform infrared (FTIR) images. The fused images show minimal distortion and the higher spatial resolution of the H&E images overcomes the diffraction limit on the spatial resolution of the FTIR images. A consideration of the FTIR spectra of nucleic acids and collagen can explain the changes in contrast between non-cancerous oral epithelium and underlying stroma within fused images formed by combining an H&E stain of oral tissue with FTIR images of the tissue obtained at a number of wavenumbers.

A novel machine learning algorithm is shown to accurately discriminate between oral squamous cell carcinoma (OSCC) nodal metastases and surrounding lymphoid tissue on the basis of a single metric, the ratio of Fourier transform infrared (FTIR) absorption intensities at 1252 cm-1 and 1285 cm-1. The metric yields discriminating sensitivities, specificities and precision of 98.8 ± 0.1%, 99.89 ± 0.01% and 99.78 ± 0.02% respectively, and an area under receiver operator characteristic (AUC) of 0.9935 ± 0.0006. The delineation of the OSCC and lymphoid tissue revealed by the image formed from the metric is in better agreement with an immunohistochemistry (IHC) stained image than are either of the FTIR images obtained at the individual wavenumbers. Scanning near-field optical microscopy (SNOM) images of the tissue obtained at a number of key wavenumbers, with high spatial resolution, show variations in the chemical structure of the tissue with a feature size down to ∼4 μm. The image formed from the ratio of the SNOM images obtained at 1252 cm-1 and 1285 cm-1 shows more contrast than the SNOM images obtained at these or a number of other individual wavenumbers. The discrimination between the two tissue types is dominated by the contribution from the 1252 cm-1 signal, which is representative of nucleic acids, and this shows the OSCC tissue to be accompanied by two wide arcs of tissue which are particularly low in nucleic acids. Haematoxylin and eosin (H&E) staining shows the tumour core in this specimen to be ∼40 μm wide and the SNOM topography shows that the core centre is raised by ∼1 μm compared to the surrounding tissue. Line profiles of the SNOM signal intensity taken through the highly keratinised core show that the increase in height correlates with an increase in the protein signal. SNOM line profiles show that the nucleic acids signal decreases at the centre of the tumour core between two peaks of higher intensity. All these nucleic acid features are ∼25 μm wide, roughly the width of two cancer cells.

A de-waxing protocol that successfully removes paraffin from tissue microarray (TMA) cores of fixed tissue obtained from oral cancer is described. The success of the protocol is demonstrated by the comparison of Fourier transform infrared (FTIR) results obtained on paraffin-embedded and de-waxed tissue and the absence of any significant correlations between infrared scanning near-field optical microscopy (SNOM) images of de-waxed tissue obtained at the three main paraffin IR peaks. The success of the protocol in removing paraffin from tissue is also demonstrated by images obtained with scanning electron microscopy (SEM) and by energy dispersive spectra (EDS) of a de-waxed CaF2 disc which shows no significant contribution from carbon. The FTIR spectra of the de-waxed TMA core overlaps that obtained from OE19 oesophageal cancer cells which had never been exposed to paraffin.

It is demonstrated that a novel multivariate analysis technique can discriminate with accuracies in the range 81-97% between Fourier transform infrared (FTIR) images of esophageal cancer OE19 and OE21 cell lines, and between esophageal cancer associated myofibroblast (CAM) and adjacent tissue myofibroblast (ATM) cells. The latter cells are morphologically indistinguishable but are known to have functionally important differences in their capacity to stimulate cancer cell growth; this report provides the first accurate spectral discrimination between CAM and ATM cells taken from the same patient. Rapid and accurate discrimination between cell types was achieved, and key wavenumbers were identified which uniquely discriminate between all four cell types. This metrics-based analysis (MA) method is shown to be unique for distinguishing between cancer stromal cells from the same patient. The key wavenumbers differ significantly from those typically found to discriminate between various esophageal cell and tissue types. A comparison is made between the MA and the established Random Forest method, and the advantages of the MA are discussed. Crucially the findings suggest a novel method that allows cancer staging based discrimination of the stromal cell types that provide the niche for tumor development.

This work reports the use of an infrared spectroscopic version of scanning near-field optical microscopy (IR-SNOM) and shows that it is possible to reveal subcellular entities via their chemical constituents in a label-free human oesophageal adenocarcinoma cell (OE33). This work presents the first high spatial resolution (∼0.15 μm) study of features imaged at various wavelengths within a cancer cell. The clear illumination of sub-micron sized moieties within a cell is a major step forward and is a key requirement for understanding cancer and for the study of other diseases and healthy tissue. The stable and tuneable light source was provided by the infrared free electron laser on the ALICE accelerator at Daresbury. The images reveal a structure with a size and wavelength absorption that are consistent with a chromosome and open the possibility of observing other localized structures, such as microvesicles, that play an important role in the development and spread of cancers.

Asingle human oesophageal adenocarcinoma cell (OE33) has been imaged using aperture infrared scanning near-field optical microscopy (IR-SNOM) in transmission and reflection and also by Fourier-transform infrared (FTIR) microspectroscopy in transmission only. This work presents the first images obtained in both transmission and reflection of the same specimen using the aperture IR-SNOM technique. The results have been used to compare the two SNOM modes and also the two techniques, which have complementary capabilities. The SNOM technique necessitates a very stable source and a careful choice of wavelengths, since it is too slow to yield images at the thousands of wavelengths obtained with FTIR. However the SNOM technique is not diffraction limited and with careful fabrication of tips can yield images with high spatial resolution. There is no significant correlation between the SNOM images obtained in transmission and reflection and the correlations between images obtained at different wavelengths vary with the different imaging modes. These results are attributed to the strong dependence of the evanescent wave on both the wavelength and the distance between the tip and the source of the signal within the sample. While both transmission and reflection SNOM images show some correlation with topography this is not a dominant effect. These results indicate that with suitable calibration a combination of reflection and transmission aperture IR-SNOM measurements has the potential to reveal information on the depth distribution of the chemical structure of a specimen.

The development of more accurate and sensitive diagnostic techniques is a key factor in efforts to improve cancer survival rates. The technique of infrared aperture fibre scanning near‐field optical microscopy (IR‐SNOM), together with radiation from the infrared free‐electron laser (IR‐FEL) on ALICE at Daresbury Laboratory (UK), has been used to obtain IR images of a crypt‐like feature and the surrounding tissue; the tissue was taken from a patient with oesophageal adenocarcinoma and with a history of Barrett's oesophagus. We have shown that the DNA signal is enhanced relative to other contributions in the region of the crypt, and the glycoprotein signal shows a less pronounced increase in the region of the crypt. The Amide II signal is found to be anti‐correlated with the DNA and glycoprotein profiles. The absorbance of the Amide II signal is found to differ for three different types of cancer tissue. High‐resolution IR images of the crypt reveal additional structure that would not be resolved in diffraction‐limited techniques.

This work reports the first images obtained by combining an infrared aperture scanning near-field optical microscope (SNOM) with a quantum cascade laser (QCL). The future potential of this set-up is demonstrated by a preliminary study on an OE33 human oesophageal adenocarcinoma cell in which the cell is imaged at 1751 cm(-1), 1651 cm(-1), 1539 cm(-1) and 1242 cm(-1). In addition to the 1651 cm(-1) image, three other images were acquired within the Amide I band (1689 cm(-1), 1675 cm(-1) and 1626 cm(-1)) chosen to correspond to secondary structures of proteins. The four images obtained within the Amide I band show distinct differences demonstrating the potential of this approach to reveal subtle changes in the chemical composition of a cell.

Lyotropic liquid crystal dispersions have commercial importance in the formulation of cleaning products and pharmaceuticals. The anisotropy of such dispersions is an important aspect of their properties and a simple method formeasurementwould be valuable to formulators. The relatively new optical technique of reflectance anisotropy spectroscopy (RAS) was considered to have potential in this respect. To test the idea, the anisotropy of a binarymixture of an anionic surfactant Aerosol OT (AOT) and glycerol (used as a model for water) was investigated using RAS. The variation in the measured anisotropy parallels the expected behaviour of a dispersion of lyotropic liquid crystals as a function of concentration and temperature. A response surface of anisotropy generated as a function of temperature and surfactant concentration demonstrates the use of RAS as a tool for mapping liquid anisotropy to facilitate the formulation of structured liquids.

A characteristic reflection anisotropy spectrum (RAS) is observed from a Au(110) surface in a wide range of electrolytes and combinations of pH and applied potentials. It is suggested that this common RAS profile arises from an interaction between the potential applied to the Au(110) electrode and the dipole moments of oxidized species that locates the Fermi level at a common position with respect to the electronic band structure of Au. Rapid changes in this RAS profile are observed for Au(110)/H2SO4 as the potential is switched between 0.3 V and 0.6 V, a potential range in which the surface is not reconstructed and below the potential range of surface oxidation. The spectral changes are completed in less than 10 ms, are reversible and are attributed to the replacement of adsorbed anions by an oxygenated species.

Bilayers and multilayers of a mutated form of cytochrome P450 reductase, P499C, form ordered structures when adsorbed at Au(110)/phosphate buffer interfaces. Reflection anisotropy spectroscopy (RAS) indicates that the degree of order increases with the number of layers and that the optical dipoles that contribute to the RAS profiles observed from the bilayer lie in planes that are oriented vertical to the surface and along one of the principal axes of the Au(110) surface. The contribution of the Au(110) substrate and the protein to the RAS profiles of the combined systems show an opposite dependence on the potential applied to the Au(110) electrode. The RAS profile observed from an adsorbed multilayer is dominated by the contribution from the protein. (C) 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

The reflection anisotropy (RAS) profiles of the Au(1 1 0)-(1  ×  1), (1  ×  2) and (1  ×  3) surface structures in electrochemical environments are shown to arise mainly from surface dipole transitions directed along the principal axes of the Au(1 1 0) surface. There are weak contributions to the RAS profiles of the Au(1 1 0)-(1  ×  1) and (1  ×  3) surfaces in the region of 4.0 eV which probably arise from (1 1 1) facets that are either intrinsic to the surface structures or are associated with steps. A transition involving a surface state just above the Fermi level, E F, contributes to the RAS profiles of the (1  ×  2) and (1  ×  3) surfaces but not to the RAS profile of the (1  ×  1) surface. A strong feature at 2.5 eV in the RAS profiles of the Au(1 1 0)-(1  ×  1) and (1  ×  2) surfaces is attributed to a transition in the vicinity of the L point of the Brillouin zone between the 5d band and the [Formula: see text] band at E F. It is argued that the applied potential of  -0.6 V, which creates the Au(1 1 0)-(1  ×  3) surface, lifts E F above the [Formula: see text] band causing it to become occupied and quenching this contribution to the RAS profile.

An analysis of the reflection anisotropy spectrum (RAS) of an ordered monolayer of cytosine adsorbed at a Au(110)/electrolyte interface is found to contain optical contributions from both the substrate and the cytosine. The spectrum of cytosine in an aqueous environment is significantly broadened by the interaction between the molecule and the Au(110). Successful simulations of the Au(110)/cytosine interface consisting of two additional molecular transitions, which sit in the middle of previously observed molecular absorption bands, are produced by an empirical Lorentzian transition model that is consistent with previous theoretical and experimental studies. While this analysis alone cannot determine the number of π→π* dipole transitions, it confirms that the only cytosine transitions that contribute to the optical response of the Au(110)/cytosine interface are located in the plane of the molecule, which is vertical to the gold surface with the long axis along the [1$\bar 1$0] direction.

Changes in the reflection anisotropy (RAS) profile of the Au(1 1 0)-(1  ×  3)/Na2SO4 interface over 25 h are attributed to the slow accumulation of impurities on the Au(1 1 0) surface which reduce the intensity of a transition involving a surface state that makes a positive contribution to the RAS profile at 1.8 eV. The growth in the intensity of a feature that makes a negative contribution to the RAS profile at 2.6 eV and the reduction in the intensity of contributions to higher energy is attributed to shifts in the energy of the surface band structure relative to the Fermi level caused by the accumulation of impurities. There is no clear explanation of the subsequent decay of the 2.6 eV feature or the long term reduction in intensity to high energy of the RAS profile.