Development of Classification Rules for a Screening Diagnostics of Lung Cancer Patients Based on the Spectral Analysis of Metabolic Profiles in the Exhaled Air

A V Borisov,V V Nikolaev,A V Shapovalov,D A Vrazhnov,Yu V Kistenev,D A Kuzmin

doi:10.1007/978-3-319-62870-7_60

A V Borisov, V V Nikolaev + Show 4 more

Open Access

https://doi.org/10.1007/978-3-319-62870-7_60

Copy DOI

Publication Date: Dec 5, 2017
Citations: 2	License type: CC BY 4.0

Affiliation: National Research Tomsk State University

Abstract

The pattern recognition technique was used for the development of classification rules for a screening diagnostics of lung cancer (LC) patients, based on the spectral analysis of metabolic profiles in the exhaled air, measured by the IR laser photoacoustic spectroscopy (LPAS). The study involved LC, chronic obstructive pulmonary disease, pneumonia patients, and healthy volunteers. The analysis of the measured spectra of exhaled air samples was based first on reduction of the dimension of the feature space using principal component analysis (PCA); thereafter the dichotomous classification was carried out using the support vector machine (SVM). The approaches to differential diagnostics based on the set of SVM classifiers usage are presented.

Full Text