Advancing cervical cancer diagnosis and screening with spectroscopy and machine learning.

INTRODUCTION: In the UK alone, the incidence of cervical cancer is increasing, hence an urgent need for early and rapid detection of cancer before it develops. Spectroscopy in conjunction with machine learning offers a disruptive technology that promises to pick up cancer early as compared to the current diagnostic techniques used. AREAS COVERED: This review article explores the different spectroscopy techniques that have been used for the analysis of cervical cancer. Along with the extensive description of spectroscopic techniques, the various machine learning techniques are also described as well as the applications that have been explored in the diagnosis of cervical cancer. This review delimits the literature specifically associated with cervical cancer studies performed solely with the use of a spectroscopy technique, and machine learning. EXPERT OPINION: Although there are several methods and techniques to detect cervical cancer, the clinical sector requires to introduce new diagnostic technologies that help improve the quality of life of patients. These innovative technologies involve spectroscopy as a qualitative method and machine learning as a quantitative method. In this article, both the techniques and methodologies that allow and promise to be a new screening tool for the detection of cervical cancer are covered.

Mid-infrared spectral classification of endometrial cancer compared to benign controls in serum or plasma samples.

This study demonstrates a discrimination of endometrial cancer versus (non-cancerous) benign controls based on mid-infrared (MIR) spectroscopy of dried plasma or serum liquid samples. A detailed evaluation was performed using four discriminant methods (LDA, QDA, kNN or SVM) to execute the classification task. The discriminant methods used in the study comprised methods that are widely used in the statistics (LDA and QDA) and machine learning literature (kNN and SVM). Of particular interest, is the impact of discrimination when presented with spectral data from a section of the bio-fingerprint region (1430 cm-1 to 900 cm-1) in contrast to the more extended bio-fingerprint region used here (1800 cm-1 to 900 cm-1). Quality metrics used were the misclassification rate, sensitivity, specificity, and Matthew's correlation coefficient (MCC). For plasma (with spectral data ranging from 1430 cm-1 to 900 cm-1), the best performing classifier was kNN, which achieved a sensitivity, specificity and MCC of 0.865 ± 0.043, 0.865 ± 0.023 and 0.762 ± 0.034, respectively. For serum (in the same wavenumber range), the best performing classifier was LDA, achieving a sensitivity, specificity and MCC of 0.899 ± 0.023, 0.763 ± 0.048 and 0.664 ± 0.067, respectively. For plasma (with spectral data ranging from 1800 cm-1 to 900 cm-1), the best performing classifier was SVM, with a sensitivity, specificity and MCC of 0.993 ± 0.010, 0.815 ± 0.000 and 0.815 ± 0.010, respectively. For serum (in the same wavenumber range), QDA performed best achieving a sensitivity, specificity and MCC of 0.852 ± 0.023, 0.700 ± 0.162 and 0.557 ± 0.012, respectively. Our findings demonstrate that even when a section of the bio-fingerprint region has been removed, good classification of endometrial cancer versus non-cancerous controls is still maintained. These findings suggest the potential of a MIR screening tool for endometrial cancer screening.

Raman spectroscopy of blood and urine liquid biopsies for ovarian cancer diagnosis: identification of chemotherapy effects.

Blood plasma and serum Raman spectroscopy for ovarian cancer diagnosis has been applied in pilot studies, with promising results. Herein, a comparative analysis of these biofluids, with a novel assessment of urine, was conducted by Raman spectroscopy application in a large patient cohort. Spectra were obtained through samples measurements from 116 ovarian cancer patients and 307 controls. Principal component analysis identified significant spectral differences between cancers without previous treatment (n = 71) and following neo-adjuvant chemotherapy (NACT), (n = 45). Application of five classification algorithms achieved up to 73% sensitivity for plasma, high specificities and accuracies for both blood biofluids, and lower performance for urine. A drop in sensitivities for the NACT group in plasma and serum, with an opposite trend in urine, suggest that Raman spectroscopy could identify chemotherapy-related changes. This study confirms that biofluids' Raman spectroscopy can contribute in ovarian cancer's diagnostic work-up and demonstrates its potential in monitoring treatment response.

Detection of ovarian cancer (± neo-adjuvant chemotherapy effects) via ATR-FTIR spectroscopy: comparative analysis of blood and urine biofluids in a large patient cohort.

Ovarian cancer remains the most lethal gynaecological malignancy, as its timely detection at early stages remains elusive. Attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy of biofluids has been previously applied in pilot studies for ovarian cancer diagnosis, with promising results. Herein, these initial findings were further investigated by application of ATR-FTIR spectroscopy in a large patient cohort. Spectra were obtained by measurements of blood plasma and serum, as well as urine, from 116 patients with ovarian cancer and 307 patients with benign gynaecological conditions. A preliminary chemometric analysis revealed significant spectral differences in ovarian cancer patients without previous chemotherapy (n = 71) and those who had received neo-adjuvant chemotherapy-NACT (n = 45), so these groups were compared separately with benign controls. Classification algorithms with blind predictive model validation demonstrated that serum was the best biofluid, achieving 76% sensitivity and 98% specificity for ovarian cancer detection, whereas urine exhibited poor performance. A drop in sensitivities for the NACT ovarian cancer group in plasma and serum indicates the potential of ATR-FTIR spectroscopy to identify chemotherapy-related spectral changes. Comparisons of regression coefficient plots for identification of biomarkers suggest that glycoproteins (such as CA125) are the main classifiers for ovarian cancer detection and responsible for smaller differences in spectra between NACT patients and benign controls. This study confirms the capacity of biofluids' ATR-FTIR spectroscopy (mainly blood serum) to diagnose ovarian cancer with high accuracy and demonstrates its potential in monitoring response to chemotherapy, which is reported for the first time. ATR-FTIR spectroscopy of blood serum achieves good segregation of ovarian cancers from benign controls, with attenuation of differences following neo-adjuvant chemotherapy.

Melanins as Sustainable Resources for Advanced Biotechnological Applications.

Melanins are a class of biopolymers that are widespread in nature and have diverse origins, chemical compositions, and functions. Their chemical, electrical, optical, and paramagnetic properties offer opportunities for applications in materials science, particularly for medical and technical uses. This review focuses on the application of analytical techniques to study melanins in multidisciplinary contexts with a view to their use as sustainable resources for advanced biotechnological applications, and how these may facilitate the achievement of the United Nations Sustainable Development Goals.

The evolving role of MUC16 (CA125) in the transformation of ovarian cells and the progression of neoplasia.

MUC16 (the cancer antigen CA125) is the most commonly used serum biomarker in epithelial ovarian cancer, with increasing levels reflecting disease progression. It is a transmembrane glycoprotein with multiple isoforms, undergoing significant changes through the metastatic process. Aberrant glycosylation and cleavage with overexpression of a small membrane-bound fragment consist MUC16-related mechanisms that enhance malignant potential. Even MUC16 knockdown can induce an aggressive phenotype but can also increase susceptibility to chemotherapy. Variable MUC16 functions help ovarian cancer cells avoid immune cytotoxicity, survive inside ascites and form metastases. This review provides a comprehensive insight into MUC16 transformations and interactions, with description of activated oncogenic signalling pathways, and adds new elements on the role of its differential glycosylation. By following the journey of the molecule from pre-malignant states to advanced stages of disease it demonstrates its behaviour, in relation to the phenotypic shifts and progression of ovarian cancer. Additionally, it presents proposed differences of MUC16 structure in normal/benign conditions and epithelial ovarian malignancy.

A comparative analysis of different biofluids towards ovarian cancer diagnosis using Raman microspectroscopy.

Biofluids, such as blood plasma or serum, are currently being evaluated for cancer detection using vibrational spectroscopy. These fluids contain information of key biomolecules, such as proteins, lipids, carbohydrates and nucleic acids, that comprise spectrochemical patterns to differentiate samples. Raman is a water-free and practically non-destructive vibrational spectroscopy technique, capable of recording spectrochemical fingerprints of biofluids with minimum or no sample preparation. Herein, we compare the performance of these two common biofluids (blood plasma and serum) together with ascitic fluid, towards ovarian cancer detection using Raman microspectroscopy. Samples from thirty-eight patients were analysed (n = 18 ovarian cancer patients, n = 20 benign controls) through different spectral pre-processing and discriminant analysis techniques. Ascitic fluid provided the best class separation in both unsupervised and supervised discrimination approaches, where classification accuracies, sensitivities and specificities above 80% were obtained, in comparison to 60-73% with plasma or serum. Ascitic fluid appears to be rich in collagen information responsible for distinguishing ovarian cancer samples, where collagen-signalling bands at 1004 cm-1 (phenylalanine), 1334 cm-1 (CH3CH2 wagging vibration), 1448 cm-1 (CH2 deformation) and 1657 cm-1 (Amide I) exhibited high statistical significance for class differentiation (P < 0.001). The efficacy of vibrational spectroscopy, in particular Raman spectroscopy, combined with ascitic fluid analysis, suggests a potential diagnostic method for ovarian cancer. Raman microspectroscopy analysis of ascitic fluid allows for discrimination of patients with benign gynaecological conditions or ovarian cancer.

A three-dimensional discriminant analysis approach for hyperspectral images.

Raman hyperspectral imaging is a powerful technique that provides both chemical and spatial information of a sample matrix being studied. The generated data are composed of three-dimensional (3D) arrays containing the spatial information across the x- and y-axis, and the spectral information in the z-axis. Unfolding procedures are commonly employed to analyze this type of data in a multivariate fashion, where the spatial dimension is reshaped and the spectral data fits into a two-dimensional (2D) structure and, thereafter, common first-order chemometric algorithms are applied to process the data. There are only a few algorithms capable of working with the full 3D array. Herein, we propose new algorithms for 3D discriminant analysis of hyperspectral images based on a three-dimensional principal component analysis linear discriminant analysis (3D-PCA-LDA) and a three-dimensional discriminant analysis quadratic discriminant analysis (3D-PCA-QDA) approach. The analysis was performed in order to discriminate simulated and real-world data, comprising benign controls and ovarian cancer samples based on Raman hyperspectral imaging, in which 3D-PCA-LDA and 3D-PCA-QDA achieved far superior performance than classical algorithms using unfolding procedures (PCA-LDA, PCA-QDA, partial lest squares discriminant analysis [PLS-DA], and support vector machines [SVM]), where the classification accuracies improved from 66% to 83% (simulated data) and from 50% to 100% (real-world dataset) after employing the 3D techniques. 3D-PCA-LDA and 3D-PCA-QDA are new approaches for discriminant analysis of hyperspectral images multisets to provide faster and superior classification performance than traditional techniques.

Detecting Endometrial Cancer by Blood Spectroscopy: A Diagnostic Cross-Sectional Study.

Endometrial cancer is the sixth most common cancer in women, with a rising incidence worldwide. Current approaches for the diagnosis and screening of endometrial cancer are invasive, expensive or of moderate diagnostic accuracy, limiting their clinical utility. There is a need for cost-effective and minimally invasive approaches to facilitate the early detection and timely management of endometrial cancer. We analysed blood plasma samples in a cross-sectional diagnostic accuracy study of women with endometrial cancer (n = 342), its precursor lesion atypical hyperplasia (n = 68) and healthy controls (n = 242, total n = 652) using attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy and machine learning algorithms. We show that blood-based infrared spectroscopy has the potential to detect endometrial cancer with 87% sensitivity and 78% specificity. Its accuracy is highest for Type I endometrial cancer, the most common subtype, and for atypical hyperplasia, with sensitivities of 91% and 100%, and specificities of 81% and 88%, respectively. Our large-cohort study shows that a simple blood test could enable the early detection of endometrial cancer of all stages in symptomatic women and provide the basis of a screening tool in high-risk groups. Such a test has the potential not only to differentially diagnose endometrial cancer but also to detect its precursor lesion atypical hyperplasia-the early recognition of which may allow fertility sparing management and cancer prevention.

Standardization of complex biologically derived spectrochemical datasets.

Spectroscopic techniques such as Fourier-transform infrared (FTIR) spectroscopy are used to study interactions of light with biological materials. This interaction forms the basis of many analytical assays used in disease screening/diagnosis, microbiological studies, and forensic/environmental investigations. Advantages of spectrochemical analysis are its low cost, minimal sample preparation, non-destructive nature and substantially accurate results. However, an urgent need exists for repetition and validation of these methods in large-scale studies and across different research groups, which would bring the method closer to clinical and/or industrial implementation. For this to succeed, it is important to understand and reduce the effect of random spectral alterations caused by inter-individual, inter-instrument and/or inter-laboratory variations, such as variations in air humidity and CO2 levels, and aging of instrument parts. Thus, it is evident that spectral standardization is critical to the widespread adoption of these spectrochemical technologies. By using calibration transfer procedures, in which the spectral response of a secondary instrument is standardized to resemble the spectral response of a primary instrument, different sources of variation can be normalized into a single model using computational-based methods, such as direct standardization (DS) and piecewise direct standardization (PDS); therefore, measurements performed under different conditions can generate the same result, eliminating the need for a full recalibration. Here, we have constructed a protocol for model standardization using different transfer technologies described for FTIR spectrochemical applications. This is a critical step toward the construction of a practical spectrochemical analysis model for daily routine analysis, where uncertain and random variations are present.

An analysis of benign human prostate offers insights into the mechanism of apocrine secretion and the origin of prostasomes.

The structure and function of normal human prostate is still not fully understood. Herein, we concentrate on the different cell types present in normal prostate, describing some previously unreported types and provide evidence that prostasomes are primarily produced by apocrine secretion. Patients (n = 10) undergoing TURP were prospectively consented based on their having a low risk of harbouring CaP. Scanning electron microscopy and transmission electron microscopy was used to characterise cell types and modes of secretion. Zinc levels were determined using Inductively Coupled Plasma Mass Spectrometry. Although merocrine secretory cells were noted, the majority of secretory cells appear to be apocrine; for the first time, we clearly show high-resolution images of the stages of aposome secretion in human prostate. We also report a previously undescribed type of epithelial cell and the first ultrastructural image of wrapping cells in human prostate stroma. The zinc levels in the tissues examined were uniformly high and X-ray microanalysis detected zinc in merocrine cells but not in prostasomes. We conclude that a significant proportion of prostasomes, possibly the majority, are generated via apocrine secretion. This finding provides an explanation as to why so many large proteins, without a signal peptide sequence, are present in the prostatic fluid.

A three-dimensional principal component analysis approach for exploratory analysis of hyperspectral data: identification of ovarian cancer samples based on Raman microspectroscopy imaging of blood plasma.

Hyperspectral imaging is a powerful tool to obtain both chemical and spatial information of biological systems. However, few algorithms are capable of working with full three-dimensional images, in which reshaping or averaging procedures are often performed to reduce the data complexity. Herein, we propose a new algorithm of three-dimensional principal component analysis (3D-PCA) for exploratory analysis of complete 3D spectrochemical images obtained through Raman microspectroscopy. Blood plasma samples of ten patients (5 healthy controls, 5 diagnosed with ovarian cancer) were analysed by acquiring hyperspectral imaging in the fingerprint region (∼780-1858 cm-1). Results show that 3D-PCA can clearly differentiate both groups based on its scores plot, where higher loadings coefficients were observed in amino acids, lipids and DNA regions. 3D-PCA is a new methodology for exploratory analysis of hyperspectral imaging, providing fast information for class differentiation.

Blood-based near-infrared spectroscopy for the rapid low-cost detection of Alzheimer's disease.

Alzheimer's disease (AD) is currently under-diagnosed and is predicted to affect a great number of people in the future, due to the unrestrained aging of the population. An accurate diagnosis of AD at an early stage, prior to (severe) symptomatology, is of crucial importance as it would allow the subscription of effective palliative care and/or enrolment into specific clinical trials. Today, new analytical methods and research initiatives are being developed for the on-time diagnosis of this devastating disorder. During the last decade, spectroscopic techniques have shown great promise in the robust diagnosis of various pathologies, including neurodegenerative diseases and dementia. In the current study, blood plasma samples were analysed with near-infrared (NIR) spectroscopy as a minimally-invasive method to distinguish patients with AD (n = 111) from non-demented volunteers (n = 173). After applying multivariate classification models (principal component analysis with quadratic discriminant analysis - PCA-QDA), AD individuals were correctly identified with 92.8% accuracy, 87.5% sensitivity and 96.1% specificity. Our results show the potential of NIR spectroscopy as a simple and cost-effective diagnostic tool for AD. Robust and early diagnosis may be a first step towards tackling this disease by allowing timely intervention.

Raman spectroscopic techniques to detect ovarian cancer biomarkers in blood plasma.

Robust diagnosis of ovarian cancer is crucial to improve patient outcomes. The lack of a single and accurate diagnostic approach necessitates the advent of novel methods in the field. In the present study, two spectroscopic techniques, Raman and surface-enhanced Raman spectroscopy (SERS) using silver nanoparticles, have been employed to identify signatures linked to cancer in blood. Blood plasma samples were collected from 27 patients with ovarian cancer and 28 with benign gynecological conditions, the majority of which had a prolapse. Early ovarian cancer cases were also included in the cohort (n = 17). The derived information was processed to account for differences between cancerous and healthy individuals and a support vector machine (SVM) algorithm was applied for classification. A subgroup analysis using CA-125 levels was also conducted to rule out that the observed segregation was due to CA-125 differences between patients and controls. Both techniques provided satisfactory diagnostic accuracy for the detection of ovarian cancer, with spontaneous Raman achieving 94% sensitivity and 96% specificity and SERS 87% sensitivity and 89% specificity. For early ovarian cancer, Raman achieved sensitivity and specificity of 93% and 97%, respectively, while SERS had 80% sensitivity and 94% specificity. Five spectral biomarkers were detected by both techniques and could be utilised as a panel of markers indicating carcinogenesis. CA-125 levels did not seem to undermine the high classification accuracies. This minimally invasive test may provide an alternative diagnostic and screening tool for ovarian cancer that is superior to other established blood-based biomarkers.

Potential of mid-infrared spectroscopy as a non-invasive diagnostic test in urine for endometrial or ovarian cancer.

The current lack of an accurate, cost-effective and non-invasive test that would allow for screening and diagnosis of gynaecological carcinomas, such as endometrial and ovarian cancer, signals the necessity for alternative approaches. The potential of spectroscopic techniques in disease investigation and diagnosis has been previously demonstrated. Here, we used attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy to analyse urine samples from women with endometrial (n = 10) and ovarian cancer (n = 10), as well as from healthy individuals (n = 10). After applying multivariate analysis and classification algorithms, biomarkers of disease were pointed out and high levels of accuracy were achieved for both endometrial (95% sensitivity, 100% specificity; accuracy: 95%) and ovarian cancer (100% sensitivity, 96.3% specificity; accuracy 100%). The efficacy of this approach, in combination with the non-invasive method for urine collection, suggest a potential diagnostic tool for endometrial and ovarian cancers.

Raman Spectroscopy to Diagnose Alzheimer's Disease and Dementia with Lewy Bodies in Blood.

Accurate identification of Alzheimer's disease (AD) is still of major clinical importance considering the current lack of noninvasive and low-cost diagnostic approaches. Detection of early stage AD is particularly desirable as it would allow early intervention or recruitment of patients into clinical trials. There is also an unmet need for discrimination of AD from dementia with Lewy bodies (DLB), as many cases of the latter are misdiagnosed as AD. Biomarkers based on a simple blood test would be useful in research and clinical practice. Raman spectroscopy has been implemented to analyze blood plasma of a cohort that consisted of early stage AD, late-stage AD, DLB, and healthy controls. Classification algorithms achieved high accuracy for the different groups: early stage AD vs healthy with 84% sensitivity, 86% specificity; late-stage AD vs healthy with 84% sensitivity, 77% specificity; DLB vs healthy with 83% sensitivity, 87% specificity; early-stage AD vs DLB with 81% sensitivity, 88% specificity; late-stage AD vs DLB with 90% sensitivity, 93% specificity; and lastly, early-stage AD vs late-stage AD 66% sensitivity and 83% specificity. G-score values were also estimated between 74% and 91%, demonstrating that the overall performance of the classification model was satisfactory. The wavenumbers responsible for differentiation were assigned to important biomolecules, which can serve as a panel of biomarkers. These results suggest a cost-effective, blood-based test for neurodegeneration in dementias.

Synchrotron- and focal plane array-based Fourier-transform infrared spectroscopy differentiates the basalis and functionalis epithelial endometrial regions and identifies putative stem cell regions of human endometrial glands.

The cyclical process of regeneration of the endometrium suggests that it may contain a cell population that can provide daughter cells with high proliferative potential. These cell lineages are clinically significant as they may represent clonogenic cells that may also be involved in tumourigenesis as well as endometriotic lesion development. To determine whether the putative stem cell location within human uterine tissue can be derived using vibrational spectroscopy techniques, normal endometrial tissue was interrogated by two spectroscopic techniques. Paraffin-embedded uterine tissues containing endometrial glands were sectioned to 10-µm-thick parallel tissue sections and were floated onto BaF2 slides for synchrotron radiation-based Fourier-transform infrared (SR-FTIR) microspectroscopy and globar focal plane array-based FTIR spectroscopy. Different spectral characteristics were identified depending on the location of the glands examined. The resulting infrared spectra were subjected to multivariate analysis to determine associated biophysical differences along the length of longitudinal and crosscut gland sections. Comparison of the epithelial cellular layer of transverse gland sections revealed alterations indicating the presence of putative transient-amplifying-like cells in the basalis and mitotic cells in the functionalis. SR-FTIR microspectroscopy of the base of the endometrial glands identified the location where putative stem cells may reside at the same time pointing towards νsPO2- in DNA and RNA, nucleic acids and amide I and II vibrations as major discriminating factors. This study supports the view that vibration spectroscopy technologies are a powerful adjunct to our understanding of the stem cell biology of endometrial tissue. Graphical abstract ᅟ.

Vibrational spectroscopy: a promising approach to discriminate neurodegenerative disorders.

Neurodegenerative diseases are a growing burden in modern society, thus crucially calling for the development of accurate diagnostic strategies. These diseases are currently incurable, a fact which has been attributed to their late diagnosis, after brain damage has already become widespread. An earlier and improved diagnosis is necessary for the enrolment of patients into clinical trials and can pave the way for the development of therapeutic tactics. Novel analytical techniques, such as mass spectrometry and vibrational spectroscopy, have been able to successfully detect and characterise neurodegenerative disorders. It is critical to globally support and make use of innovative basic research and techniques, which could ultimately lead to the creation of a cost-effective diagnostic test. Minimally invasive samples, such as biological fluids, have also been shown to reveal information for these diseases; utilising them could simplify sample collection/analysis and be more preferable to patients.

Aluminium foil as an alternative substrate for the spectroscopic interrogation of endometrial cancer.

Biospectroscopy has the potential to investigate and characterize biological samples and could, therefore, be utilized to diagnose various diseases in a clinical environment. An important consideration in spectrochemical studies is the cost-effectiveness of the substrate used to support the sample, as high expense would limit their translation into clinic. In this paper, the performance of low-cost aluminium (Al) foil substrates was compared with the commonly used low-emissivity (low-E) slides. Attenuated total reflection-Fourier transform infrared spectroscopy was used to analyse blood plasma and serum samples from women with endometrial cancer and healthy controls. The 2 populations were differentiated using principal component analysis with support vector machines with 100% sensitivity in plasma samples (endometrial cancer = 70; healthy controls = 15) using both Al foil and low-E slides as substrates. The same sensitivity results (100%) were achieved for serum samples (endometrial cancer = 60; healthy controls = 15). Specificity was found higher using Al foil (90%) in comparison to low-E slides (85%) and lower using Al foil (70%) in comparison to low-E slides in serum samples. The establishment of Al foil as low-cost and highly performing substrate would pave the way for large-scale, multicentre studies and potentially for routine clinical use.