Optical pathology using oral tissue fluorescence spectra: classification by principal component analysis and k-means nearest neighbor analysis

Comparing Preoperative Screening Tools for Elective Urologic Cancer Surgery: Insights from a Cluster Analysis

INTRODUCTION

The aim of this study was to evaluate the features and benefits of different geriatric screening tools for enhancing the perioperative care of patients who undergo elective cancer surgery using cluster analysis.

METHODS

This study was a retrospective, observational analysis of 1,019 consecutive patients who had elective major cancer surgery in the urology department of our hospital from October 2019 to January 2023. Before the surgery, a trained nurse screened the patients using six tools: Eastern Clinical Oncology Group performance status (ECOG-PS), flemish version of the triage risk screening tool (fTRST), geriatric-8 (G8), instrumental activities of daily living, patient health questionnaire-2 (PHQ-2), and simple questionnaire to rapidly diagnose sarcopenia (SARC-F). The study grouped the patients into four clusters based on their scores on these tools and compared their outcomes after the surgery. The outcomes included overall survival, ambulation failure, delirium, and severe complications. The study also examined how each screening tool was associated with the outcomes.

RESULTS

Based on their clinical data and screening results, we classified the patients into four groups: Healthy (73%), Depressive (11%), Intermediate (11%), and Unhealthy (5%). The Unhealthy group had the worst outcomes in overall survival (OS), ambulation failure, and delirium, followed by the Intermediate group. In addition, fTRST and SARC-F emerged as significant predictors of OS; ECOG-PS, fTRST, G8, and SARC-F of ambulation failure; ECOG-PS, fTRST, and G8 of delirium; and G8 of severe complications.

CONCLUSION

Various geriatric screening tools were found to have the potential to forecast diverse postoperative outcomes.

Diffuse Reflectance Spectroscopy of the Oral Mucosa: In Vivo Experimental Validation of the Precancerous Lesions Early Detection Possibility

This article is devoted to the experimental validation of the possibility of early detection of precancerous lesions in the oral mucosa in vivo using diffuse reflectance spectroscopy in the wavelength range from 360 to 1000 nm. During the study, a sample of 119 patients with precancerous lesions has been collected and analyzed. As a result of the analysis, the most informative wavelength ranges were determined, in which the maximum differences in the backscattering spectra of lesions and intact tissues were observed, methods for automatic classification of backscattering spectra of the oral mucosa were studied, sensitivity and specificity values, achievable using diffuse reflectance spectroscopy for detecting hyperkeratosis on the tongue ventrolateral mucosa surface and buccal mucosa, were evaluated. As a result of preliminary experimental studies in vivo, the possibility of automatic detection of precancerous lesions of the oral mucosa surface using diffuse reflectance spectroscopy in the wavelength range from 500 to 900 nm with an accuracy of at least 75 percent has been shown.

Au/SiNCA-based SERS analysis coupled with machine learning for the early-stage diagnosis of cisplatin-induced liver injury

Cisplatin has been widely applied in the clinical treatment of various cancers, whereas liver injury induced by its hepatotoxicity is still a severe issue. Reliable identification of early-stage cisplatin-induced liver injury (CILI) can improve clinical care and help to streamline drug development. Traditional methods, however, cannot achieve enough information at the subcellular level due to the requirement of the labeling process and low sensitivity. To overcome these, we designed an Au-coated Si nanocone array (Au/SiNCA) to fabricate the microporous chip as the surface-enhanced Raman scattering (SERS) analysis platform for the early diagnosis of CILI. A CILI rat model was established, and the exosome spectra were obtained. The principal component analysis (PCA)-representation coefficient-based k-nearest centroid neighbor (RCKNCN) classification algorithm was proposed as the multivariate analysis method to build the diagnosis and staging model. The PCA-RCKNCN model has been validated to achieve a satisfactory result, with accuracy and AUC of over 97.5%, and sensitivity and specificity of over 95%, indicating that SERS combined with the PCA-RCKNCN analysis platform can be a promising tool for clinical applications.

Construction of machine learning-based models for cancer outcomes in low and lower-middle income countries: A scoping review

Background

The impact and utility of machine learning (ML)-based prediction tools for cancer outcomes including assistive diagnosis, risk stratification, and adjunctive decision-making have been largely described and realized in the high income and upper-middle-income countries. However, statistical projections have estimated higher cancer incidence and mortality risks in low and lower-middle-income countries (LLMICs). Therefore, this review aimed to evaluate the utilization, model construction methods, and degree of implementation of ML-based models for cancer outcomes in LLMICs.

Methods

PubMed/Medline, Scopus, and Web of Science databases were searched and articles describing the use of ML-based models for cancer among local populations in LLMICs between 2002 and 2022 were included. A total of 140 articles from 22,516 citations that met the eligibility criteria were included in this study.

Results

ML-based models from LLMICs were often based on traditional ML algorithms than deep or deep hybrid learning. We found that the construction of ML-based models was skewed to particular LLMICs such as India, Iran, Pakistan, and Egypt with a paucity of applications in sub-Saharan Africa. Moreover, models for breast, head and neck, and brain cancer outcomes were frequently explored. Many models were deemed suboptimal according to the Prediction model Risk of Bias Assessment tool (PROBAST) due to sample size constraints and technical flaws in ML modeling even though their performance accuracy ranged from 0.65 to 1.00. While the development and internal validation were described for all models included (n=137), only 4.4% (6/137) have been validated in independent cohorts and 0.7% (1/137) have been assessed for clinical impact and efficacy.

Conclusion

Overall, the application of ML for modeling cancer outcomes in LLMICs is increasing. However, model development is largely unsatisfactory. We recommend model retraining using larger sample sizes, intensified external validation practices, and increased impact assessment studies using randomized controlled trial designs

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=308345, identifier CRD42022308345.

Accuracy of artificial intelligence-assisted detection of Oral Squamous Cell Carcinoma: A systematic review and meta-analysis

Oral Squamous Cell Carcinoma (OSCC) is an aggressive tumor with a poor prognosis. Accurate and timely diagnosis is therefore essential for reducing the burden of advanced disease and improving outcomes. In this meta-analysis, we evaluated the accuracy of artificial intelligence (AI)-assisted technologies in detecting OSCC. We included studies that validated any diagnostic modality that used AI to detect OSCC. A search was performed in six databases: PubMed, Embase, Scopus, Cochrane Library, ProQuest, and Web of Science up to 15 Mar 2022. The Quality Assessment Tool for Diagnostic Accuracy Studies was used to evaluate the included studies' quality, while the Split Component Synthesis method was utilized to quantitatively synthesize the pooled diagnostic efficacy estimates. We considered 16 out of the 566 yielded studies, which included twelve different AI models with a total of 6606 samples. The summary sensitivity, summary specificity, positive and negative likelihood ratios as well as the pooled diagnostic odds ratio were 92.0 % (95 % confidence interval [CI] 86.7-95.4 %), 91.9 % (95 % CI 86.5-95.3 %), 11.4 (95 % CI 6.74-19.2), 0.087 (95 % CI 0.051-0.146) and 132 (95 % CI 62.6-277), respectively. Our findings support the capability of AI-assisted systems to detect OSCC with high accuracy, potentially aiding the histopathological examination in early diagnosis, yet more prospective studies are needed to justify their use in the real population.

Non-invasive imaging of oral potentially malignant and malignant lesions: A systematic review and meta-analysis

Non-invasive (NI) imaging techniques have been developed to overcome the limitations of invasive biopsy procedures, which is the gold standard in diagnosis of oral dysplasia and Oral Squamous Cell Carcinoma (OSCC). This systematic review and meta- analysis was carried out with an aim to investigate the efficacy of the NI-imaging techniques in the detection of dysplastic oral potentially malignant disorders (OPMDs) and OSCC. Records concerned in the detection of OPMDs, Oral Cancer were identified through search in PubMed, Science direct, Cochrane Library electronic database (January 2000 to October 2020) and additional manual searches. Out of 529 articles evaluated for eligibility, 56 satisfied the pre-determined inclusion criteria, including 13 varying NI-imaging techniques. Meta-analysis consisted 44 articles, wherein majority of the studies reported Autofluorescence (AFI-38.6%) followed by Chemiluminescence (CHEM), Narrow Band Imaging (NBI) (CHEM, NBI-15.9%), Fluorescence Spectroscopy (FS), Diffuse Reflectance Spectroscopy (DRS), (FS, DRS-13.6%) and 5aminolevulinic acid induced protoporphyrin IX fluorescence (5ALA induced PPIX- 6.8%). Higher sensitivities (Sen) and specificities (Spe) were obtained using FS (Sen:74%, Spe:96%, SAUC=0.98), DRS (Sen:79%, Spe:86%, SAUC = 0.91) and 5 ALA induced PPIX (Sen:91%, Spe:78%, SAUC = 0.98) in the detection of dysplastic OPMDs from non-dysplastic lesions(NDLs). AFI, FS, DRS, NBI showed higher sensitivities and SAUC (>90%) in differentiating OSCC from NDLs. Analysed NI-imaging techniques suggests the higher accuracy levels in the diagnosis of OSCC when compared to dysplastic OPMDs. 5 ALA induced PPIX, DRS and FS showed evidence of superior accuracy levels in differentiation of dysplastic OPMDs from NDLs, however results need to be validated in a larger number of studies.

Skin optical properties in the obese and their relation to body mass index: a review

SIGNIFICANCE

Obesity is a worldwide epidemic contributing directly to several cardiovascular risk factors including hypertension and type 2 diabetes. Wearable devices are becoming better at quantifying biomarkers relevant for the management of health and fitness. Unfortunately, both anecdotal evidence and recent studies indicate that some wearables have higher levels of error when utilized by populations with darker skin tones and high body mass index (BMI). There is an urgent need for a better evaluation of the limits of wearable health technologies when used by obese individuals.

AIMS

(1) To review the current know-how on changes due to obesity in the skin epidermis, dermis, and subcutis that could affect the skin optical properties; (2) for the green wavelength range, to evaluate the difference in absorption and scattering coefficients from the abdominal skin between individuals with and without elevated BMI. The changes include alterations in layer thickness and cell size, as well as significant differences in chromophores and scatterer content, e.g., water, hemoglobin, collagen, and lipids.

APPROACH

We have summarized literature pertaining to changes in skin and its components in obesity and report the results of our search using articles published between years 1971 and 2020. A linear model was used to demonstrate the absorption and reduced scattering coefficient of the abdominal skin of individuals with and without elevated BMI in the green wavelength range (530 to 550 nm) that is typically found in most wearables.

RESULTS

The general trends indicate a decrease in absorption for both dermis and subcutis and an increase in reduced scattering for both epidermis and dermis. At 544-nm wavelength, a typical wavelength used for photoplethysmography (PPG), the absorption coefficient's relative percentage difference between high and low BMI skin, was 49% in the subcutis, 19% in the dermis, and negligible in the epidermis, whereas the reduced scattering coefficient relative difference was 21%, 29%, and 165% respectively.

CONCLUSIONS

These findings suggest that there could be significant errors in the output of optical devices used for monitoring health and fitness if changes due to obesity are not accounted for in their design.

Detection of mitochondrial dysfunction in vitro by laser-induced autofluorescence

Mitochondrion plays a significant role in a variety of biological functions. Because of their diverse character and location in the cellular systems, mitochondria commonly get exposed to various extrinsic and intrinsic cellular stresses. The present study reports a novel approach to detection of mitochondrial dysfunction based on tryptophan autofluorescence of its proteins in mouse liver, using laser-induced fluorescence (LIF) as a tool. Mitochondria, isolated from the mouse liver, were initially tested for purity and integrity using lactate dehydrogenase and succinate dehydrogenase (SDH) assays. Mitochondrial stress was induced by treating the isolated mitochondria with heavy metals at 10 and 0.01 mM for sodium arsenite and mercuric chloride, respectively. Upon treatment with the heavy metal, tryptophan autofluorescence quenching was recorded at 281 nm excitation. The functional integrity of the mitochondria treated with heavy metals was evaluated by measuring SDH and cytochrome c oxidase activities at various concentrations of mitochondria, which showed impaired activity as compared to control upto a concentration of 6.25 μg. A significant shift was also observed in the autofluorescence of proteins upto the level below 1 μg, suggesting their conformational change and hence altered structural integrity of mitochondria. Circular dichroism spectroscopy data of the mitochondrial proteins treated with heavy metals further validates their conformational change as compared to untreated control. The present study clearly shows that the LIF can be a novel detection tool to detect altered structural integrity of cellular mitochondria upon stress, and it also possesses the potentiality to combine with other interdisciplinary modalities.

Probing endogenous collagen by laser-induced autofluorescence in burn wound biopsies: A pilot study

The focus of the current study was to interrogate the predictive potential of laser-induced autofluorescence (LIAF) by objectively assessing collagen synthesis in burn wound granulation tissues ex vivo. Prior grafting, granulation tissues (20 samples) following burn injury were collected from 17 subjects of age range 18 to 60 years with patient/donor consent and the corresponding autofluorescence spectra were recorded at 325 nm He-Cd laser (≈2 mW) excitations. The resulting endogenous collagen intensity from the above tissue samples was computed by normalizing the nicotinamide adenine dinucleotide levels. In addition, the hydroxyproline content was also estimated biochemically from the same granulation tissues. A comparative assessment of both LIAF and biochemical estimations for endogenous collagen by hydroxyproline resulted in strong positive correlation among them. The above relevant observations suggest that LIAF is equally informative as that of biochemical estimations, in evaluating endogenous collagen content in wound granulation tissues. Thus, it can be concluded that LIAF has the predictive potential, as a noninvasive objective tool to measure the endogenous collagen levels in wound biopsy tissues and provide complementary data conducive for making clinical decisions.

Spatio-temporal map for early cancer detection: Proof of concept

A spatio-temporal map of human cervical tissue is obtained from time-resolved fluorescence images with the dynamic contrast enhanced through principal component analysis (PCA) for clear demarcation of regions of normal and pre-cancerous conditions. Changes in the properties of fluorescence in different environments are captured through fluorescence lifetime maps in the human cervical tissue sample. The correlation embodied in the second principal component (PC) representing sectorial information free of background of the first PC, segregates fluorescence activities, as illustrated in the PC maps. It significantly enhances the contrast of the images which are majorly handicapped by the variations in fluorophore environment. The result is validated on phantoms, mimicking the changes in the environment of normal and abnormal tissues. This spatio-temporal map illustrates the potential of time resolved auto-fluorescence imaging of cervical tissue in combination with PCA to clearly demarcate normal and abnormal regions with enhanced contrast.

Assessment of the sensitivity and specificity of tissue-specific-based and anatomical-based optical biomarkers for rapid detection of human head and neck squamous cell carcinoma

OBJECTIVES

We propose the use of morphological optical biomarkers for rapid detection of human head and neck squamous cell carcinoma (HNSCC) by leveraging the underlying tissue characteristics in aerodigestive tracts.

MATERIALS AND METHODS

Diffuse reflectance spectra were obtained from malignant and contra-lateral normal tissues of 57 patients undergoing panendoscopy and biopsy. Oxygen saturation, total hemoglobin concentration, and the reduced scattering coefficient were extracted. Differences in malignant and normal tissues were examined based on two different groupings: anatomical site and morphological tissue type.

RESULTS AND CONCLUSIONS

Measurements were acquired from 252 sites, of which 51 were pathologically classified as SCC. Optical biomarkers exhibited statistical differences between malignant and normal samples. Contrast was enhanced when parsing tissues by morphological classification rather than anatomical subtype for unpaired comparisons. Corresponding linear discriminant models using multiple optical biomarkers showed improved predictive ability when accounting for morphological classification, particularly in node-positive lesions. The false-positive rate was retrospectively found to decrease by 34.2% in morphologically- vs. anatomically-derived predictive models. In glottic tissue, the surgeon exhibited a false-positive rate of 45.7% while the device showed a lower false-positive rate of 12.4%. Additionally, comparisons of optical parameters were made to further understand the physiology of tumor staging and potential causes of high surgeon false-positive rates. Optical spectroscopy is a user-friendly, non-invasive tool capable of providing quantitative information to discriminate malignant from normal head and neck tissues. Predictive models demonstrated promising results for real-time diagnostics. Furthermore, the strategy described appears to be well suited to reduce the clinical false-positive rate.

Objective assessment of endogenous collagen in vivo during tissue repair by laser induced fluorescence

Collagen, a triple helical protein with the primary role of mechanical function, provides tensile strength to the skin, and plays a pivotal task in tissue repair. During tissue regeneration, collagen level increases gradually and therefore, monitoring of such changes in vivo by laser induced fluorescence was the main objective behind the present study. In order to accomplish this, 15 mm diameter excisional wounds were created on six to eight week old Swiss albino mice. The collagen deposition accelerated upon irradiation of single exposure of 2 J/cm² He-Ne laser dose immediately after wounding was recorded by laser induced autofluorescence in vivo along with un-illuminated and un-wounded controls. Autofluorescence spectra were recorded for each animal of the experimental groups on 0, 5, 10, 30, 45 and 60 days post-wounding, by exciting the granulation tissue/skin with 325 nm He-Cd laser. The variations in the average collagen intensities from the granulation tissue/skin of mice were inspected as a function of age and gender. Further, the spectral findings of the collagen synthesis in wound granulation tissue/un-wounded skin tissues were validated by Picro-Sirius red- polarized light microscopy in a blinded manner through image analysis of the respective collagen birefringence. The in vivo autofluorescence studies have shown a significant increase in collagen synthesis in laser treated animals as compared to the un-illuminated controls. Image analysis of the collagen birefringence further authenticated the ability of autofluorescence in the objective monitoring of collagen in vivo. Our results clearly demonstrate the potential of laser induced autofluorescence in the monitoring of collegen synthesis during tissue regeneration, which may have clinical implications.

Tissue discrimination by uncorrected autofluorescence spectra: a proof-of-principle study for tissue-specific laser surgery

Laser surgery provides a number of advantages over conventional surgery. However, it implies large risks for sensitive tissue structures due to its characteristic non-tissue-specific ablation. The present study investigates the discrimination of nine different ex vivo tissue types by using uncorrected (raw) autofluorescence spectra for the development of a remote feedback control system for tissue-selective laser surgery. Autofluorescence spectra (excitation wavelength 377 ± 50 nm) were measured from nine different ex vivo tissue types, obtained from 15 domestic pig cadavers. For data analysis, a wavelength range between 450 nm and 650 nm was investigated. Principal Component Analysis (PCA) and Quadratic Discriminant Analysis (QDA) were used to discriminate the tissue types. ROC analysis showed that PCA, followed by QDA, could differentiate all investigated tissue types with AUC results between 1.00 and 0.97. Sensitivity reached values between 93% and 100% and specificity values between 94% and 100%. This ex vivo study shows a high differentiation potential for physiological tissue types when performing autofluorescence spectroscopy followed by PCA and QDA. The uncorrected autofluorescence spectra are suitable for reliable tissue discrimination and have a high potential to meet the challenges necessary for an optical feedback system for tissue-specific laser surgery.

Support Vector Machine on fluorescence landscapes for breast cancer diagnostics

Excitation-emission matrices (EEM) and total synchronous fluorescence spectra (SFS) of normal and malignant breast tissue specimens are measured in UV-VIS spectral region to serve as data inputs in development of Support Vector Machine (SVM) based breast cancer diagnostics tool. Various input data combinations are tested for classification accuracy using SVM prediction against histopathology findings to discover the best combination regarding diagnostics sensitivity and specificity. It is shown that with EEM data SVM provided 67% sensitivity and 62% specificity diagnostics. With SFS data SVM provided 100% sensitivity and specificity for a several input data combinations. Among these combinations those that require minimal data inputs are identified.

Spectroscopic and histological evaluation of wound healing progression following Low Level Laser Therapy (LLLT)

The present study focuses on the evaluation of the effect of He-Ne laser on tissue regeneration by monitoring collagen synthesis in wound granulation tissues in Swiss albino mice using analysis of laser induced fluorescence (LIF) and light microscopy techniques. The spectral analyses of the wound granulation tissues have indicated a dose dependent increase in collagen levels during the post-wounding days. The histological examinations on the other hand have also shown a significant increase in collagen deposition along with the reduced edema, leukocytes, increased granulation tissue, and fibroblast number in the optimal laser dose treated group compared to the non-illuminated controls.

Autofluorescence spectroscopy for evaluating dysplasia in colorectal tissues

The aim of this study was to assess the applicability of autofluorescence (AF) spectroscopy as a method for the diagonosis of normal, benign and malignant of dysplasia in colorectal tissues experimentally. By improvement of optical design in laser pulse generator, wavelength-adjustable output was acquired and the optimal wavelength was defined as 380 nm. With 380-nm pulsed laser excitation, AF spectra of normal, benign and malignant colorectal tissues were recorded in the spectra region from 460-570 nm in vitro. The spectral analysis for discrimination among the different types of tissues was carried out using principal component analysis (PCA)-based Neural networks algorithm. The performance of analysis was pretty good with sensitivity, specificity and accuracy found to be 100%,90% and 96.7%, respectively. The AF spectroscopy may serve as an excellent tool for the evaluation of dysplasia in colorectal clinical diagnosis.

Comparative evaluation of the diagnostic performance of autofluorescence and diffuse reflectance in oral cancer detection: a clinical study

Autofluorescence (AF) and diffuse reflectance (DR) spectroscopic techniques have shown good diagnostic accuracies for noninvasive detection of oral cavity cancer. In the present study, AF and DR spectra recorded in vivo from the same set of sites in 65 patients were analyzed using Principal component analysis (PCA) and linear discriminant analysis (LDA). The effectiveness of these two techniques was assessed by comparison with gold standard and their discrimination efficiency was determined from the area under the receiver operator characteristic (AUC-ROC) curve. Analysis using a DR technique shows a higher AUC-ROC of 0.991 as against 0.987 for AF spectral data.

Photoacoustic spectroscopy of ovarian normal, benign, and malignant tissues: a pilot study

Photoacoustic spectra of normal, benign, and malignant ovarian tissues are recorded using 325-nm pulsed laser excitation in vitro. A total of 102 (34 normal, 38 benign, and 30 malignant) spectra are obtained from 22 samples belonging to normal, benign, and malignant subjects. Applying multi-algorithm approach, comprised of methods such as, principal component analysis (PCA) based k-nearest neighbor (k-NN) analysis, artificial neural network (ANN) analysis, and support vector machine (SVM) analysis, classification of the data has been carried out. For PCA, first the calibration set is formed by pooling 45 spectra, 15 belonging to each of pathologically certified normal, benign, and malignant samples. PCA is then performed on the data matrix, comprised of the six spectral features extracted from each of 45 calibration samples, and three principal components (PCs) containing maximum diagnostic information are selected. The scores of the selected PCs are used to train the k-NN, ANN, and SVM classifiers. The ANN used is a classical multilayer feed forward network with back propagation algorithm for its training. For k-NN, the Euclidean distance based algorithm is used and for SVM, one-versus-rest multiclass kernel-radial basis function is used. The performance evaluation of the classification results are obtained by calculating statistical parameters like specificity and sensitivity. ANN and k-NN techniques showed identical performance with specificity and sensitivity values of 100 and 86.76%, whereas SVM had these values at 100 and 80.18%, respectively. In order to determine the relative diagnostic performance of the techniques, receiver operating characteristics analysis is also performed.

Autofluorescence of osteoporotic mouse femur bones: a pilot study

OBJECTIVE

The objective of this study was to evaluate the efficacy of laser-induced fluorescence (LIF) in combination with principal component analysis (PCA) to characterize osteoporotic conditions induced by ovariectomy (OVX) in mice.

BACKGROUND DATA

The dual energy X-ray absorptiometry is the gold standard methodology routinely used to diagnose osteoporosis. In recent years, the use of LIF to characterize human disease and to aid in diagnosis has shown great promise. However, this technique has not been much exploited for monitoring osteoporosis.

MATERIALS AND METHODS

Swiss albino mice were sacrificed 2, 3, 4, and 5 wk after OVX and their femur bones were excised. The same protocol was used for age-matched female controls (no OVX). The LIF spectra from different regions of the bones were recorded and compared using PCA.

RESULTS

A significant change in the fluorescence pattern of osteoporotic bones compared with the control was indicated by PCA match/no-match analysis. A region-wise PCA match/no-match analysis of the spectral changes against respective region calibration sets indicated more no-matches in the fifth week bones compared with the others. Further, the spectral differences were more prominent in the proximal and distal parts of the bones. In addition to the PCA, Gaussian curve fitting was also performed on control, third week, and fifth week bone spectra to identify different spectral components in them.

CONCLUSION

This preliminary study using fluorescence spectroscopy in combination with PCA clearly demonstrated osteoporotic changes in mouse femur bones at different time points after OVX, suggesting possible human applications.

Adaptive spectral window sizes for extraction of diagnostic features from optical spectra

We present an approach to adaptively adjust the spectral window sizes for optical spectra feature extraction. Previous studies extracted features from spectral windows of a fixed width. In our algorithm, piecewise linear regression is used to adaptively adjust the window sizes to find the maximum window size with reasonable linear fit with the spectrum. This adaptive windowing technique ensures the signal linearity in defined windows; hence, the adaptive windowing technique retains more diagnostic information while using fewer windows. This method was tested on a data set of diffuse reflectance spectra of oral mucosa lesions. Eight features were extracted from each window. We performed classifications using linear discriminant analysis with cross-validation. Using windowing techniques results in better classification performance than not using windowing. The area under the receiver-operating-characteristics curve for windowing techniques was greater than a nonwindowing technique for both normal versus mild dysplasia (MD) plus severe high-grade dysplasia or carcinama (SD) (MD+SD) and benign versus MD+SD. Although adaptive and fixed-size windowing perform similarly, adaptive windowing utilizes significantly fewer windows than fixed-size windows (number of windows per spectrum: 8 versus 16). Because adaptive windows retain most diagnostic information while reducing the number of windows needed for feature extraction, our results suggest that it isolates unique diagnostic features in optical spectra.

Principal component analysis (PCA)-based k-nearest neighbor (k-NN) analysis of colonic mucosal tissue fluorescence spectra

OBJECTIVE

The objective of this study was to verify the suitability of principal component analysis (PCA)-based k-nearest neighbor (k-NN) analysis for discriminating normal and malignant autofluorescence spectra of colonic mucosal tissues.

BACKGROUND DATA

Autofluorescence spectroscopy, a noninvasive technique, has high specificity and sensitivity for discrimination of diseased and nondiseased colonic tissues. Previously, we assessed the efficacy of the technique on colonic data using PCA Match/No match and Artificial Neural Networks (ANNs) analyses. To improve the classification reliability, the present work was conducted using PCA-based k-NN analysis and was compared with previously obtained results.

METHODS

A total of 115 fluorescence spectra (69 normal and 46 malignant) were recorded from 13 normal and 10 malignant colonic tissues with 325 nm pulsed laser excitation in the spectral region 350-600 nm in vitro. We applied PCA to extract the relevant information from the spectra and used a nonparametric k-NN analysis for classification.

RESULTS

The normal and malignant spectra showed large variations in shape and intensity. Statistically significant differences were found between normal and malignant classes. The performance of the analysis was evaluated by calculating the statistical parameters specificity and sensitivity, which were found to be 100% and 91.3%, respectively.

CONCLUSION

The results obtained in this study showed good discrimination between normal and malignant conditions using PCA-based k-NN analysis.

Synchronous fluorescence spectroscopy for the detection and characterization of cervical cancers in vitro

The objective of this study was to assess the diagnostic potential of synchronous fluorescence (SF) spectroscopy (SFS) technique for the detection and characterization of normal and different malignancy stages of moderately differentiated squamous cell carcinoma (MDSCC), poorly differentiated squamous cell carcinoma (PDSCC) cervical tissues. SF spectra were measured from 45 biopsies from 30 patients in vitro. Characteristic, highly resolved peaks and significant spectral differences between normal and MDSCC, PDSCC cervical tissues were obtained. Nine potential ratios were calculated and used as input variables for a discriminant analysis across different groups. The potentiality of the SFS technique was estimated by two discriminant analyses. Discriminant analysis I performed across normal and abnormal (including MDSCC and PDSCC) cervical tissues classified as 100% both original and the cross-validated grouped cases. In discriminant analysis II performed across the three groups, normal, MDSCC and PDSCC, 100% of both original and the cross-validated grouped cases were correctly classified. Using the SFS technique, one can obtain all the key biochemical markers such as tryptophan, collagen, hemoglobin, reduced form of nicotinamide adenine dinucleotide and flavin adenine dinucleotide in a single scan and hence they can be targeted as tumor markers in the detection of normal from abnormal cervical tissues.

Classification of ultraviolet irradiated mouse skin histological stages by bimodal spectroscopy: multiple excitation autofluorescence and diffuse reflectance

Histopathological analysis and in vivo optical spectroscopy were used to discriminate several histological stages of UV-irradiated mouse skin. At different times throughout the 30-week irradiation, autofluorescence (AF) and diffuse reflectance (DR) spectra were acquired in a bimodal approach. Then skin was sampled and processed to be classified, according to morphological criteria, into four histological categories: normal, and three types of hyperplasia (compensatory, atypical, and dysplastic). After extracting spectral characteristics, principal component analysis (data reduction) and the k-nearest neighbor classifying method were applied to compare diagnostic performances of monoexcitation AF (based on each of the seven excitation wavelengths: 360, 368, 390, 400, 410, 420, and 430 nm), multiexcitation AF (combining the seven excitation wavelengths), DR, and bimodal spectroscopies. Visible wavelengths are the most sensitive ones to discriminate compensatory from precancerous (atypical and dysplastic) states. Multiexcitation AF provides an average 6-percentage-point increased sensitivity compared to the best scores obtained with monoexcitation AF for all pairs of tissue categories. Bimodality results in a 4-percentage-point increase of specificity when discriminating the three types of hyperplasia. Thus, bimodal spectroscopy appears to be a promising tool to discriminate benign from precancerous stages; clinical investigations should be carried out to confirm these results.