Monday , 09 September 2013


HALL 1-41         Session 272        12:50-14:40  


Thematic Poster Session : New methods for diagnostic workup of lung cancer  

Analysis of exhaled breath with electronic nose and diagnosis of lung cancer by multifactorial logistic regression analysis

M. Bukovskis, G. Strazda, N. Jurka, U. Kopeika, A. Pirtnieks, L. Balode, J. Aprinceva, I. Kantane, I. Taivans (Riga, Latvia)

Exhaled breath of lung cancer patients contains specific pattern of volatile organic compounds (VOCs).
The aim of our study was to test the potential of multifactorial logistic regression (MLRA) analysis in diagnosis of lung cancer.
Exhaled 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. Exhaled air was collected using standardized method and sampled by electronic nose (Cyranose 320). Optimal detector parameter combination and methematical model for discrimination of lung cancer was calculated by MLRA backward stepwise method. Sensitivity, specificity, positive (PPV) and negative predictive value (NPV) of the method in the training group of smokers and nonsmokers was calculated.
Total 475 patients, out of them 252 lung cancer patients and 223 patients with different lung diseases and healthy volunteers, and 265 current nonsmokers and 210 smokers, were recruited in the study.
Classification of cases in nonsmokers
 Lung cancerNo cancer 
Lung cancer12812PPV 91.4%
No cancer5120NPV 96.0%
 Sensitivity 96.2%Specificity 90.6% 

Classification of cases in smokers
 Lung cancerNo cancer 
Lung cancer1147PPV 94,2%
No cancer584NPV 94.4%
 Sensitivity 95.8%Specificity 92.3% 
Finding of optimal detector parameter combination and division of patients in smokers and nonsmokers give very high lung cancer prediction accuracy with MLRA.
Study was sponsored by ERAF activity Project Nb. 2010/0303/2DP/