Effectiveness of lung cancer early diagnosis by analysing exhaled breath composition using neural network and multimodal approach

Background. The five-year survival rate of lung cancer patients remains extremely low, with the average rate of 22 %. Early detection of this disease can improve survival rates and reduce mortality. Lung cancer development is influenced by various risk factors, with smoking being the most significant, followed by other factors like occupational exposure, infections, genetic predisposition, presence of chronic diseases, etc. Considering risk factors is crucial for improving the efficacy of automated gas analysis complexes for lung cancer diagnosis. These complexes are promising for the application of scalable neural network data processing algorithms for noninvasive diagnosis of early stage lung cancer.

Purpose of the study: to evaluate the effectiveness of the multimodal lung cancer detection method by analyzing the exhaled breath composition from 100 volunteers using simultaneous assessment of the exhaled breath composition and risk factors.

Material and Methods. Along with exhaled breath samples, the study database also recorded all volunteers’ medical history data. Neural networks with a variety of architectures were used for data processing. The dataset for training neural network classifiers included exhaled breath samples from 100 volunteers, including 47 from healthy subjects and 53 from patients with morphologically confirmed lung cancer. Data determining the age group, smoking status and the presence of chronic lung diseases were analyzed as risk factors for lung cancer.

Results. The integration of data, including exhaled breath composition and lung cancer risk factors, into a single neural network results in a 3 % increase in its original monomodal architecture. Taking into account a relatively small number of risk factors, such as age, gender, smoking status and COPD, increases the classifier’s sensitivity by 1.89 % and specificity by 6.39 %. The best generalization performance of the neural network classifier is achieved with a two-stream hybrid model with normalization.

Conclusion. By incorporating diverse patient history data, such as age, smoking history, and chronic diseases, in addition to exhaled air composition data, into a unified neural network classifier, the accuracy of the exhaled air method for lung cancer diagnosis can be increased by an average of 4 %. This improvement, coupled with the expanded range of risk factors considered in the future application of the exhaled air method for lung cancer diagnosis in medical practice, will improve the reliability of population screening results.

1. Adib E., Nassar A.H., Abou Alaiwi S., Groha S., Akl E.W., Sholl L.M., Michael K.S., Awad M.M., Jӓnne P.A., Gusev A., Kwiatkowski D.J. Variation in targetable genomic alterations in non-small cell lung cancer by genetic ancestry, sex, smoking history, and histology. Genome Med. 2022; 14(1): 39. doi: 10.1186/s13073-022-01041-x.

2. Schabath M.B., Cote M.L. Cancer Progress and Priorities: Lung Cancer. Cancer Epidemiol Biomarkers Prev. 2019; 28(10): 1563–79. doi: 10.1158/1055-9965.EPI-19-0221.

3. Malhotra J., Malvezzi M., Negri E., La Vecchia C., Boffetta P. Risk factors for lung cancer worldwide. Eur Respir J. 2016; 48(3): 889–902. doi: 10.1183/13993003.00359-2016.

4. LoPiccolo J., Gusev A., Christiani D.C., Jänne P.A. Lung cancer in patients who have never smoked – an emerging disease. Nat Rev Clin Oncol. 2024; 21(2): 121–46. doi: 10.1038/s41571-023-00844-0.

5. Wang W., Dou S., Dong W., Xie M., Cui L., Zheng C., Xiao W. Impact of COPD on prognosis of lung cancer: from a perspective on disease heterogeneity. Int J Chron Obstruct Pulmon Dis. 2018; 13: 3767–76. doi: 10.2147/COPD.S168048.

6. Forder A., Zhuang R., Souza V.G.P., Brockley L.J., Pewarchuk M.E., Telkar N., Stewart G.L., Benard K., Marshall E.A., Reis P.P., Lam W.L. Mechanisms Contributing to the Comorbidity of COPD and Lung Cancer. Int J Mol Sci. 2023; 24(3): 2859. doi: 10.3390/ijms24032859.

7. Park H.Y., Kang D., Shin S.H., Yoo K.H., Rhee C.K., Suh G.Y., Kim H., Shim Y.M., Guallar E., Cho J., Kwon O.J. Chronic obstructive pulmonary disease and lung cancer incidence in never smokers: a cohort study. Thorax. 2020; 75(6): 506–509. doi: 10.1136/thoraxjnl-2019-213732.

8. Wasswa-Kintu S., Gan W.Q., Man S.F., Pare P.D., Sin D.D. Relationship between reduced forced expiratory volume in one second and the risk of lung cancer: a systematic review and meta-analysis. Thorax. 2005. 60(7): 570–75. doi: 10.1136/thx.2004.037135.

9. Lam V.K., Bentzen S.M., Mohindra P., Nichols E.M., Bhooshan N., Vyfhuis M., Scilla K.A., Feigenberg S.J., Edelman M.J., Feliciano J.L. Obesity is associated with long-term improved survival in definitively treated locally advanced non-small cell lung cancer (NSCLC). Lung Cancer. 2017; 104: 52–57. doi: 10.1016/j.lungcan.2016.11.017.

10. Huang S.X., Jaurand M.C., Kamp D.W., Whysner J., Hei T.K. Role of mutagenicity in asbestos fiber-induced carcinogenicity and other diseases. J Toxicol Environ Health B Crit Rev. 2011; 14(1-4): 179–245. doi: 10.1080/10937404.2011.556051.

11. Kulbakin D.E., Obkhodskaya E.V., Obkhodskiy A.V., Rodionov E.O., Sachkov V.I., Chernov V.I., Choynzonov E.L. Study of the effectiveness of diagnostic method for respiratory system diseases by analyzing the exhaled air using a gas analytical complex. Siberian Journal of Clinical and Experimental Medicine. 2023; 38(4): 260–69. (in Russian). doi: 10.29001/2073-8552-2023-653. EDN: SRPDUH.

12. Obkhodskiy A.V., Kulbakin D.E., Obkhodskaya E.V., Popov A.S., Rodionov E.O., Sachkov V.I., Chernov V.I., Choynzonov E.L. Neural network classification software for the gas analytical survey data of respiratory system. Tomsk State University Journal of Control and Computer Science. 2024; 69: 112–23.(in Russian). doi: 10.17223/19988605/69/12. EDN: SJAGNG.

13. Ortiz B.L., Gupta V., Kumar R., Jalin A., Cao X., Ziegenbein C., Singhal A., Tewari M., Choi S.W. Data Preprocessing Techniques for AI and Machine Learning Readiness: Scoping Review of Wearable Sensor Data in Cancer Care. JMIR Mhealth Uhealth. 2024; 12: e59587. doi: 10.2196/59587.

14. Cos H., Li D., Williams G., Chininis J., Dai R., Zhang J., Srivastava R., Raper L., Sanford D., Hawkins W., Lu C., Hammill C.W. Predicting Outcomes in Patients Undergoing Pancreatectomy Using Wearable Technology and Machine Learning: Prospective Cohort Study. J Med Internet Res. 2021; 23(3): e23595. doi: 10.2196/23595.

15. Adeoye J., Su Y. Deep learning with data transformation improves cancer risk prediction in oral precancerous conditions. Intelligent Medicine. 2025; 5(2): 141–50. doi: 10.1016/j.imed.2024.11.003.

16. Papadimitroulas P., Brocki L., Christopher Chung N., Marchadour W., Vermet F., Gaubert L., Eleftheriadis V., Plachouris D., Visvikis D., Kagadis G.C., Hatt M. Artificial intelligence: Deep learning in oncological radiomics and challenges of interpretability and data harmonization. Phys Med. 2021; 83: 108–21. doi: 10.1016/j.ejmp.2021.03.009.

17. Yang T.Y., Kuo P.Y., Huang Y., Lin H.W., Malwade S., Lu L.S., Tsai L.W., Syed-Abdul S., Sun C.W., Chiou J.F. Deep-Learning Approach to Predict Survival Outcomes Using Wearable Actigraphy Device Among End-Stage Cancer Patients. Front Public Health. 2021; 9: 730150. doi: 10.3389/fpubh.2021.730150.

18. Huang Y., Roy N., Dhar E., Upadhyay U., Kabir M.A., Uddin M., Tseng C.L., Syed-Abdul S. Deep Learning Prediction Model for Patient Survival Outcomes in Palliative Care Using Actigraphy Data and Clinical Information. Cancers (Basel). 2023; 15(8): 2232. doi: 10.3390/cancers15082232.

19. Li Y., Jiang J., Li X., Zhang M. Multi-modal data integration of dosiomics, radiomics, deep features, and clinical data for radiation-induced lung damage prediction in breast cancer patients. J Radiat Research and Appl Sci. 2025; 18(2): 101389. doi: 10.1016/j.jrras.2025.101389.

20. Batra U., Nathany S., Nath S.K., Jose J.T., Sharma T., Preeti. P., Pasricha S., Sharma M., Arambam N., Khanna V., Bansal A., Mehta A., Rawal K. AI-based pipeline for early screening of lung cancer: integrating radiology, clinical, and genomics data. Lancet Reg Health Southeast Asia. 2024; 24: 100352. doi: 10.1016/j.lansea.2024.100352.