Hybrid feature selection and SVM-based classification for mouse skin precancerous stages diagnosis from bimodal spectroscopy

Faiza, Abdat; Amouroux, Marine; Guermeur, Yann; Blondel, Walter

doi:10.1364/oe.20.000228

Cited by 27 publications

(19 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These hypervolumes can be specific to the possible pathological nature of the tissue. This can be achieved through the use of data processing algorithms, such as the Kernel Principal Component Analysis or the Support Vector Machine (Diaz-Ayil, 2007); (Adbat, 2012). This hyperspectral technique can be further developed using images obtained with different excitation wavelengths.…”

Section: Possibilities With Hyperspectral Fluorescencementioning

confidence: 99%

Reverse Translational Research - How Clinical Trials on Fluorescence Imaging for Vocal Cord Cancer Fuels Fundamental Research

Gaiffe

Pieralli

Tavernier³

et al. 2013

Proceedings of the International Conference on Biomedical Electronics and Devices

View full text Add to dashboard Cite

Translational research consists in translating fundamental research results as closely as possible to patients. Researchers sometimes underestimate these studies because it is thought that, although essential for setting up new investigation tools, they do not deepen fundamental knowledge. However, users face specific difficulties due to the variability of the biological systems under study. Variability is easily understood from one patient to another, but there is also variability in a single patient whose metabolism evolves together with therapeutic actions. Results obtained in translational research often depend on this variability, and new questions and scientific obstacles arise when research is applied to the real world. In order to address these new challenges, reverse translational research is required. Fundamental research is fuelled by the results of translational research. In this position paper, we consider vocal cord fluorescence imaging as an example of bi-directional translational research. First, we briefly recall the basics of fluorescence imaging, and we explain why commercial fluorescence systems lead to variable estimations of their efficiency by end-users. Second, we describe solutions intended to improve fluorescence techniques. This position paper will then make conclusions.

show abstract

Section: Possibilities With Hyperspectral Fluorescencementioning

confidence: 99%

Reverse Translational Research - How Clinical Trials on Fluorescence Imaging for Vocal Cord Cancer Fuels Fundamental Research

Gaiffe

Pieralli

Tavernier³

et al. 2013

Proceedings of the International Conference on Biomedical Electronics and Devices

View full text Add to dashboard Cite

show abstract

“…The characteristics of the data to be classified and the type of the classification problem will determine the optimal kernel and parameters to be used. SVMs combined with medical spectroscopy samples have been widely used to analyze different types of diseases, such us the tissue characterization or the diagnostics of lymph nodes in breast cancer [11,12], the diagnosis of skin cancer in mice [13], or the analysis of blood samples to detect dengue infection [14], among others. SVMs have been also used to classify the identify samples of primary tumors of brain metastases obtained using Raman spectroscopy [15].…”

Section: Introductionmentioning

confidence: 99%

SVM Optimization for Brain Tumor Identification Using Infrared Spectroscopic Samples

Fabelo

Ortega

Casselden

et al. 2018

Sensors

View full text Add to dashboard Cite

The work presented in this paper is focused on the use of spectroscopy to identify the type of tissue of human brain samples employing support vector machine classifiers. Two different spectrometers were used to acquire infrared spectroscopic signatures in the wavenumber range between 1200–3500 cm−1. An extensive analysis was performed to find the optimal configuration for a support vector machine classifier and determine the most relevant regions of the spectra for this particular application. The results demonstrate that the developed algorithm is robust enough to classify the infrared spectroscopic data of human brain tissue at three different discrimination levels.

show abstract

“…Consequently, the spectral data variability can be important and might compromise this kind of optical biopsy. More advanced and multimodal techniques have also been proposed to more completely characterize the tissue under investigation: spatially resolved autofluorescence (13) and/or reflectance measurements (14) enable to determine the tissue physiologic state more accurately. Moreover, confocal endomicroscopy (15), based on high spatial resolution imaging of parietal structures, reveals valuable information for gastrointestinal tract cancer diagnosis in clinical centers that can afford such an investment.…”

Section: Introductionmentioning

confidence: 99%

Autofluorescence spectroscopy for multimodal tissues characterization in colitis-associated cancer murine model

Dorez¹,

Sablong²,

Canaple³

et al. 2015

SPIE Proceedings

View full text Add to dashboard Cite

International audienceThe purpose of this research project is to assess mice colon wall, using three optical modalities (conventional endoscopy, confocal endomicroscopy and optical spectroscopy) and endoluminal MRI. The study is done in the context of inflammatory bowel disease and colorectal cancer that represent 13% of new cases of cancer, every year in western countries. An optical spectroscopic bench (autofluorescence and reflectance) was developed with a flexible fiber probe. This latter has been combined with a mini multi-purpose rigid endoscope and a confocal endomicroscope. The optical modalities were first used in vivo on SWISS mice. Then, a specific optical a phantom (containing two layers of distinct fluorophores) was developed in order to evaluate our two-channel spectroscopic probe as a basic depth-sensitive measurement tool. The preliminary results show the feasibility to combine such modalities in the same in vivo protocol. Conventional endoscopy is useful to depict inflammation along colon wall. Confocal endomicroscopy provides high-contrasted images of microvascularization. Measured optical spectra both depend on biochemical tissue content and layered structure of the medium. The light collected from one channel is not similar to the other, in terms of intensity and spectroscopic profile as the interaction with the medium observed volume is different. A comparative analysis of the spectra based on our in vitro model exhibits a strong correlation between simple index extracted from spectral data and two main phantom characteristics (fluorophore concentrations and superficial layer thickness). This work suggests that this technique could contribute to assess tissues alterations through autofluorescence spectroscopic measurement under endoscopy

show abstract

Hybrid feature selection and SVM-based classification for mouse skin precancerous stages diagnosis from bimodal spectroscopy

Cited by 27 publications

References 36 publications

Reverse Translational Research - How Clinical Trials on Fluorescence Imaging for Vocal Cord Cancer Fuels Fundamental Research

Reverse Translational Research - How Clinical Trials on Fluorescence Imaging for Vocal Cord Cancer Fuels Fundamental Research

SVM Optimization for Brain Tumor Identification Using Infrared Spectroscopic Samples

Autofluorescence spectroscopy for multimodal tissues characterization in colitis-associated cancer murine model

Contact Info

Product

Resources

About