Training and Validating a Portable Electronic Nose for Lung Cancer Screening

Area of Science:

• Biomedical Engineering
• Oncology
• Analytical Chemistry

Background:

• Volatile organic compounds (VOCs) in exhaled breath are potential biomarkers for cancer diagnosis.
• Electronic nose (e-nose) technology offers a noninvasive method for VOC profiling.

Purpose of the Study:

• To evaluate the capability of a portable point-of-care e-nose to distinguish lung cancer patients from healthy individuals.
• To assess the diagnostic accuracy of the e-nose in a real-world setting.

Main Methods:

• Five e-nose devices were used to collect breath samples from 60 lung cancer patients and 107 healthy controls.
• Breath samples were analyzed to identify VOC patterns.
• An artificial neural network model was trained and validated using a blinded control group.

Main Results:

• The e-nose achieved a diagnostic accuracy of 83% (sensitivity 83%, specificity 84%) in differentiating lung cancer patients from controls.
• Validation in a blinded group yielded comparable results with 86% accuracy (sensitivity 88%, specificity 86%).

Conclusions:

• The portable e-nose demonstrates feasibility for noninvasive lung cancer detection.
• Further research with larger cohorts, including early-stage tumors, is recommended to enhance robustness and clinical utility.