Figure - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
SVM model results using linear, polynomial and Gaussian radial basis kernel functions for training data set.
Source publication
We report a significant improvement in the diagnosis of cervical cancer through a combined application of principal component analysis (PCA) and support vector machine (SVM) on the average fluorescence decay profile of Fluorescence Lifetime Images (FLI) of epithelial hyperplasia (EH) and CIN-I cervical tissue samples, obtained ex-vivo. The fast and...
Contexts in source publication
Context 1
... PC score classification, first two PCs are considered as they carry more than 99% of the original information (as seen in Figure 3a) while 3rd PC does not carry any significant information. The overall model results for PC scores using different kernels for training data is shown in Table 1. The performance of a model is generally evaluated in terms of accuracy, precision, sensitivity, and specificity. ...Context 2
... performance of a model is generally evaluated in terms of accuracy, precision, sensitivity, and specificity. From Table 1 it can be seen that polynomial and RBF kernel function performances are similar with accuracy, precision, sensitivity and specificity of 84%, 100%, 100% and 100% respectively. The higher specificity and sensitivity for non-linear SVM clearly indicates that the boundary separating CIN-I and EH samples are not linear. ...Similar publications
In this paper, we study and analyze five different approaches for gender classification (Male or Female). The purpose of approaches is to extract the main features from face image and for each approach; we used these features as input to the Support Vector Machine SVM classifier for classification process. In the first approach, we implemented the...
In real-time scenarios, tracking and monitoring a human behavior from a video is very active research. In human computer interaction, recognition and monitoring of end user's expressions, emotions and behavior from the real-time video streaming plays very important role. In such systems it is required to track the dynamic changes in human face move...
Following paper evaluates the features of an image using Wavelet Transform by using a Discrete Wavelet Transform (DWT). Principal Component Analysis (PCA) is applied to reduce the feature what is extracted. The concept of K-fold stratified cross-acceptance is used to increase conception of Kernel Support Vector Machine (KSVM). The techniques from p...
Control of the roll gap of the caster segment is one of the key parameters for ensuring the quality of a slab in continuous casting. In order to improve the precision and timeliness of the roll gap value control, we proposed a rolling gap value prediction (RGVP) method based on the continuous casting process parameters. The process parameters colle...
The ability of obtain accurate information on future river flow is a fundamental key for water resources planning, and management. Traditionally, single models have been introduced to predict the future value of river flow. This paper investigates the ability of Principal Component Analysis as dimensionality reduction technique and combined with si...
Citations
... pre-cancerous cervical tissue. [168] Tissue-engineered 3D models of biological systems are greatly advantaged by the ability to undertake non-destructive live cell imaging in order to undertake drug and toxicology screening which require repeated measures of cell function over time. To this end, one group developed a perfusion bioreactor system interfaced with a non-invasive optical imaging system which assessed NAD(P)H and FAD + autofluorescence over time in order to monitor the redox ratio of engineered 3D human adipose tissue as it developed. ...
Autofluorophores are endogenous fluorescent compounds that naturally occur in the intra and extracellular spaces of all tissues and organs. Most have vital biological functions – like the metabolic cofactors NAD(P)H and FAD⁺, as well as the structural protein collagen. Others are considered to be waste products – like lipofuscin and advanced glycation end products – which accumulate with age and are associated with cellular dysfunction. Due to their natural fluorescence, these materials have great utility for enabling non‐invasive, label‐free assays with direct ties to biological function. Numerous technologies, with different advantages and drawbacks, are applied to their assessment, including fluorescence lifetime imaging microscopy, hyperspectral microscopy, and flow cytometry. Here, the applications of label‐free autofluorophore assessment are reviewed for clinical and health‐research applications, with specific attention to biomaterials, disease detection, surgical guidance, treatment monitoring, and tissue assessment – fields that greatly benefit from non‐invasive methodologies capable of continuous, in vivo characterization.
... Among the optical techniques, fluorescence-based devices (spectroscopic and imaging) are extensively used by many researches for in vivo detection of oral cancer [18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33]. Other alternatives such as body fluids (saliva, blood, urine, etc.) have also been studied for the detection of oral mucosal lesions by the research groups [34][35][36][37]. Various biomarkers or fluorophore molecules are found in human oral cavity tissue. ...
... b Photograph of a device with the required accessories such as a hand-held device, laser diode, spectrometer, and laptop, while collecting the data from an oral carcinoma patient vectors (data points closest to the decision boundary). In the SVM classifier, the most widely used kernels such as linear, radial basis function (RBF), polynomial, and sigmoid were included, and diagnostic parameters such as accuracy, sensitivity, and specificity were computed [37][38][39][40]. All the steps included for the classifying of three groups are summarized in the flow chart and displayed in Fig. 2. ROC is used to find the discrimination line and to compute the diagnostic parameters. ...
Detection of oral mucosal lesions has been performed by an in-house developed fluorescence-based portable device in the present study. A laser diode of 405 nm wavelength and a UV-visible spectrometer are utilized in the portable device as excitation and detection sources. At the 405 nm excitation wavelength, the flavin adenine dinucleotide (FAD) band at 500 nm and three porphyrin bands at 634, 676, and 703 nm are observed in the fluorescence spectrum of the oral cavity tissue. We have conducted this clinical study on a total of 189 tissue sites of 36 oral squamous cell carcinoma (OSCC) patients, 18 dysplastic (precancerous) patients, and 34 volunteers. Analysis of the fluorescence data has been performed by using the principal component analysis (PCA) method and support vector machine (SVM) classifier. PCA is applied first in the spectral data to reduce the dimension, and then classification among the three groups has been executed by employing the SVM. The SVM classifier includes linear, radial basis function (RBF), polynomial, and sigmoid kernels, and their classification efficacies are computed. Linear and RBF kernels on the testing data sets differentiated OSCC and dysplasia to normal with an accuracy of 100% and OSCC to dysplasia with an accuracy of 95% and 97%, respectively. Polynomial and sigmoid kernels showed less accuracy values among the groups ranging from 48 to 88% and 51 to 100%, respectively. The result indicates that fluorescence spectroscopy and the SVM classifier can help to identify early oral mucosal lesions with significant high accuracy.
... In fluorescence spectroscopy, various native fluorophores like NADH, FAD, collagen, and porphyrins, present in different layers of tissue, are targeted with a particular excitation wavelength (ultraviolet or visible) and their respective fluorescence emission is observed in the long wavelength region (visible or near-infrared) [32,33]. The biochemical and morphological changes occurring in the layered tissue structure with the progression of abnormality [34][35][36][37] lead to changes in fluorescence spectral characteristics of different fluorophores. The native fluorescence of the fluorophores is also modulated by various wavelength dependent absorption and scattering present inside the biological tissue. ...
Cervical cancer is one of the most prevalent forms of cancer, with a lengthy latent period and a gradual onset phase. Conventional techniques are found to be severely lacking in real time detection of disease progression which can greatly enhance the cure rate. Due to their high sensitivity and specificity, optical techniques are emerging as reliable tools, particularly in case of cancer. It has been seen that biochemical changes are better highlighted through intrinsic fluorescence devoid of interference from absorption and scattering. Its effectiveness in in‐vivo conditions is affected by the fact that the intrinsic spectral signatures vary from patient to patient, as well as in different population groups. Here, we overcome this limitation by collectively enumerating the subtle changes in the spectral profiles and correlations through an information theory based entropic approach, which significantly amplifies the minute spectral variations. In conjunction with artificial intelligence (AI)/ machine learning (ML) tools, it yields high specificity and sensitivity with a small dataset from patients in clinical conditions, without artificial augmentation. We have used an in‐house developed handheld probe (i‐HHP) for extracting intrinsic fluorescence spectra of human cervix from 110 different subjects drawn from diverse population groups. The average classification accuracy of the proposed methodology using 10‐fold cross validation is 93.17%. A combination of polarised fluorescence spectra from i‐HHP and the proposed classifier is proven to be minimally invasive with the ability to diagnose patients in real time. This paves the way for effective use of relatively smaller sized sensitive fluorescence data with advanced AI/ML tools for early cervical cancer detection in clinics.
This article is protected by copyright. All rights reserved.
... To reach this level optical techniques have been very important for researchers and technology developers because it is based in the analyses of light as a result of its interaction with matter, e.g. reflection, transmission, absorption, refraction and diffraction [5][6][7][8][9]. ...
This paper presents new perspectives of photonic technologies for capsule endoscopy. It is firstly presented a review about conventional endoscopy (upper endoscopy and colonoscopy), followed by capsule endoscopy (CE), as well as their techniques, advantages, and drawbacks. The technologies for CEs presented in this paper have the potential of integration on the existing endoscopic systems commercially available. Such technologies include narrow band imaging (NBI), photodynamic therapy (PDT), confocal laser endomicroscopy (CLE), Optical coherence tomography (OCT) and Spectroscopy in order to improve the performance of the gastrointestinal (GI) tract examination. In the context of NBI, two optical filters were designed and fabricated for integration into endoscopic capsules, allowing the visualization of light centered at the 415 nm and 540 nm. These optical filters are based on the principle of Fabry-Perot and were made of thin-films of titanium dioxide (TiO2) and silicon dioxide (SiO2). Moreover, strategies and aid solutions for the adaptation of ECs for PDT are also discussed.
... LDA has also been extended to include kernel for nonlinear classification but the approach is again limited by the small sample size (SSS) problem. Previous studies have reported that MLAs like DTC and SVM perform better than LDA as classifier by resolving the SSS (Oliveira et al. 2005;Sahoo et al. 2018) problem. ...
Early detection of oral cancer is essential for improving survival rates and diagnostic precision. Raman cyto-spectroscopy is a useful method, but its interpretive complexity prevents frequent clinical use. Prediction algorithms have been used to evaluate RS data and create CAD systems. An ensemble of MLAs and dimensionality reduction techniques has been demonstrated to effectively analyse the complex RS. In present study we utilised two MLA ensembles—(1) Linear Discriminant Analysis with Decision Tree Classifier (2) LDA with Support Vector Machine. Results show that RS of oral epithelial cells, collected from different study groups (viz. Normal-healthy volunteers; histo-pathologically confirmed- potentially cancerable- leucoplakia and cancer patients-OSCC) when classified using the ensembles had better overall accuracy of (88 ± 8%) as compared to either DTC (48 ± 19%) or SVM (58 ± 15%) when applied individually. Class-wise evaluation showed improved performance of LDA-SVM compared to LDA-DTC, and blind test was performed with RS of habitual cigarette smokers (CS) as susceptible group and healthy volunteers as control group to assess the classifier ensembles' performance in cancer risk prediction. The LDA-SVM ensemble showed the highest classification accuracy, indicating that it can be used to accurately detect oral cancer risk and screen susceptible patients.
... To reach this level optical techniques have been very important for researchers and technology developers because it is based in the analyses of light as a result of its interaction with matter, e.g. reflexion, transmission, absorption, refraction and diffraction [5][6][7][8][9]. ...
This paper presents new perspectives of photonic technologies for capsule endoscopy. It is firstly presented a review about conventional endoscopy (upper endoscopy and colonoscopy), followed by capsule endoscopy (CE), as well as their techniques, advantages, and drawbacks. The technologies for CEs presented in this paper have the potential of integration on the existing endoscopic systems commercially available. Such technologies include narrow band imaging (NBI), photodynamic therapy (PDT), confocal laser endomicroscopy (CLE), Optical coherence tomography (OCT) and Spectroscopy in order to improve the performance of the GI tract examination. In the context of NBI, two optical filters were designed and fabricated for integration into endoscopic capsules, allowing the visualization of light centered at the 415 nm and 540 nm. These optical filters are based on the principle of Fabry-Perot and were made of thin-films of titanium dioxide (TiO2) and silicon dioxide (SiO2). Moreover, strategies and aid solutions for the adaptation of ECs for PDT are also discussed.
... Numerous papers are published annually on FLIM-assisted research on various processes in cells in vitro with both endogeneous and exogeneous fluorophores; see [22][23][24][25][26][27] for some recent examples. In vivo studies utilizing FLIM have also gained popularity both for research purposes and as a diagnostic tool in clinics [28][29][30][31]. ...
We report an investigation of the in vivo accumulation of Radachlorin photosensitizer in a murine model in several types of normal and tumor tissues based on an FLIM-assisted analysis of fluorescence intensity images, time-resolved fluorescence signals, and phasor plots. Experiments were performed on ex vivo histological samples of normal and tumor tissues. It was shown that the investigation of fluorescence intensity distributions combined with that of time-resolved fluorescence images can be used for qualitative and—under some limitations—quantitative analyses of the relative uptake of this photosensitizer in tissues. The phasor plot representations of time-resolved fluorescence signals were shown to be suitable for identification of the accumulation of predominant photosensitizers in tissues.
... Fluorescence imaging is a good technique [25][26][27][28] to visualize the difference in chlorophyll concentration in plants, so this technique would be a good choice to distinguish rapeseed grist (low chlorophyll concentration) [29,30] from the maize silage components (high chlorophyll concentration) [31,32]. The fluorescence spectra of grist and individual silage components have been acquired with the 450 nm wavelength excitation (Figure 4b). ...
A simple and cost-effective technique has been suggested for online monitoring of grist concentration in fodder. The technique is based on fluorescence imaging with grow light lamp excitation and a consumer CMOS camera (DSLR or smartphone) for photo capturing. A prototype instrument has been developed and tested in the laboratory for quantitative express determination of rapeseed grist concentration in fodder. In situ measurement of grist concentration during cattle food preparation has been demonstrated, and the perspectives were discussed. The developed instrument has the potential to ensure more accurate preparation of individual cattle diets compared to currently available methods, which will improve the efficiency of the cattle food production.
... Most recently, FLIM classifiers have been applied in vitro for label-free assessment of microglia 86 and T-cell activation, 87 as well as for exogenous labeling of intracellular components 88 and monitoring of intracellular pharmacokinetics. 89 In addition, ML classifiers have been used for FLIM-based tissue discrimination and characterization in applications including diagnosis of cervical precancer, 90 breast cancer resection 91 ...
Significance:
Biomedical optics system design, image formation, and image analysis have primarily been guided by classical physical modeling and signal processing methodologies. Recently, however, deep learning (DL) has become a major paradigm in computational modeling and has demonstrated utility in numerous scientific domains and various forms of data analysis.
Aim:
We aim to comprehensively review the use of DL applied to macroscopic diffuse optical imaging (DOI).
Approach:
First, we provide a layman introduction to DL. Then, the review summarizes current DL work in some of the most active areas of this field, including optical properties retrieval, fluorescence lifetime imaging, and diffuse optical tomography.
Results:
The advantages of using DL for DOI versus conventional inverse solvers cited in the literature reviewed herein are numerous. These include, among others, a decrease in analysis time (often by many orders of magnitude), increased quantitative reconstruction quality, robustness to noise, and the unique capability to learn complex end-to-end relationships.
Conclusions:
The heavily validated capability of DL's use across a wide range of complex inverse solving methodologies has enormous potential to bring novel DOI modalities, otherwise deemed impractical for clinical translation, to the patient's bedside.
... However, the classi er reported in this study is limited by the similar SSS problem and restrictions in input variables. Previous studies have reported that MLAs like DTC and SVM perform better than LDA as classi er by resolving SSS problem(Oliveira et al., 2005;Sahoo et al., 2018) and kernel functions are widely utilized by SVM.Present study introduces LDA as a removal of redundant features for Raman spectral analysis and couples it with DTC and SVM as classi er. This study also improvises LDA to resolve SSS problem, complexity of large feature dimension and perform non-linear classi cation. ...
Raman cyto-spectroscopy is an emerging technique for early oral cancer risk assessment. Applying machine learning algorithms (MLA) on Raman cyto-spectra of exfoliated oral epithelial cells, we present an automated and label-free method for early oral cancer risk assessment. Raman spectra were collected from healthy volunteers, patients with pre-cancerous lesions and oral sub-mucous fibrosis (OSCC). To perform nonlinear multi-class classification of Raman cyto-spectra and overcome the small sample size problem, two MLA ensembles namely linear discriminant analysis (LDA) with decision tree classifier (DTC) and LDA with support vector machine (SVM) were employed. Both LDA-DTC and LDA-SVM ensembles showed increased overall average accuracy of 88±8% compared to either DTC (48±19%) or SVM (58±15%) classifier without LDA. However, class-wise evaluation showed an improved performance of LDA-SVM compared to LDA-DTC as assessed by performance indices like sensitivity, MCC and F1 score. Further, blind test was performed with the Raman spectra of habitual cigarette smokers as susceptible group and healthy volunteers as control group for assessing the performance of the classifier ensembles in early cancer risk prediction. Misclassifications of the spectra from control group were also analysed and LDA-SVM showed highest (95%) classification accuracy. It is concluded that, LDA-SVM is a potential tool for precise detection of early oral cancer risk and can be applied to screen susceptible patients.
