RT Journal Article SR Electronic T1 Analysis of exhaled breath with electronic nose and diagnosis of lung cancer by support vector machine JF European Respiratory Journal JO Eur Respir J FD European Respiratory Society SP 1824 VO 42 IS Suppl 57 A1 Maris Bukovskis A1 Gunta Strazda A1 Normunds Jurka A1 Uldis Kopeika A1 Ainis Pirtnieks A1 Liga Balode A1 Jevgenija Aprinceva A1 Inara Kantane A1 Immanuels Taivans YR 2013 UL http://erj.ersjournals.com/content/42/Suppl_57/1824.abstract AB BackgroundExhaled breath of lung cancer patients contains unique pattern of volatile organic compounds (VOCs) which can be distinguished by analysis with electronic nose.ObjectiveWe hypothesized that electronic nose can discriminate exhaled air of patients with lung cancer from healthy controls and other lung diseases.MethodsExhaled breath of morphologically verified lung cancer patients (cancer group) and mixed group of patients with COPD, asthma, pneumonia, bronchiectasis and healthy volunteers (no cancer group) was examined. Subjects inspired VOC-filtered air by tidal breathing for 5 minutes, and a single expiratory vital capacity was collected that was sampled by electronic nose (Cyranose 320) within 5 minutes. Smellprints were analyzed by support vector machine. Age, smoking history (pack years) and ambient temperature °C were included as continuous predictors of the diagnosis. Patients were devided into 75% training and 25% test group. Cross-validation, class accuracy, sensitivity, specificity, positive (PPV) and negative predictive value (NPV) of the method was calculated in the test group.Results166 patients with lung cancer, 91 patient with other lung diseases and 79 healthy volunteers were recruited. In the test group for lung cancer cross-validation accuracy was 72.8%, class accuracy 79.1%, sensitivity 88.9%, specificity 66.7%, PPV 75.5% and NPV 83.9%.ConclusionsExhaled breath analysis by electronic nose using support vector pattern recognition method is able to discriminate lung cancer from healthy subjects and patients with different lung diseases.AcknowledgementsStudy was sponsored by ERAF activity 2.1.1.1.0 Project Nb. 2010/0303/2DP/2.1.1.1.0/10/APIA/VIAA/043/