International Validation of Electronic Nose Technology for Fibrotic Interstitial Lung Diseases
Fibrotic interstitial lung diseases, often abbreviated as fILDs, are a diverse group of lung disorders that cause scarring in the tissue around the air sacs. Over time, this scarring makes it harder for the lungs to expand and for oxygen to pass into the bloodstream. Because many forms of interstitial lung disease share similar symptoms, such as shortness of breath, dry cough, and fatigue, diagnosis can be difficult and delayed. Patients often need several tests, including lung function studies, high-resolution CT scans, blood tests, multidisciplinary review, and sometimes invasive procedures such as bronchoscopy or lung biopsy.
This international study evaluated whether electronic nose technology, also called eNose, could help distinguish between different fibrotic interstitial lung diseases using exhaled breath analysis. The results suggest that eNose testing may become a useful, noninvasive point-of-care tool to support faster diagnosis and better subtype classification.
Why Breath Analysis Matters in ILD
When a person breathes out, their breath contains hundreds of volatile organic compounds, or VOCs. These are small chemicals produced by normal metabolism, inflammation, oxidative stress, infection, and other biological processes. Different diseases can produce different VOC patterns. An electronic nose does not identify each chemical individually. Instead, it detects overall breath “fingerprints” through sensor patterns and uses statistical modeling to recognize disease-related signatures.
In interstitial lung disease, this approach is attractive because the diseases are often hard to tell apart, yet accurate subtyping is important. Some ILDs progress quickly, some respond to immunosuppressive therapy, and others may require antifibrotic treatment or close monitoring. A faster, noninvasive method could shorten the time to diagnosis and reduce the need for more invasive tests.
How the Study Was Designed
Researchers enrolled patients diagnosed with interstitial lung disease after multidisciplinary team discussion at five international expert centers. All included patients had pulmonary fibrosis seen on high-resolution chest CT scans. The study used the SpiroNose device, an electronic nose platform designed for exhaled breath analysis.
The investigators compared breath profiles in two ways. First, they tested whether each ILD subtype could be separated from all other ILDs combined. Second, they examined whether the main ILD subtypes could be distinguished from one another. The breath data were analyzed using partial least squares discriminant analysis and receiver operating characteristic, or ROC, analysis. Some centers contributed data for model training, while other centers were used for external validation, which is important because it tests whether the method still works in a different real-world setting.
Who Was Included
Breath profiles from 587 patients were analyzed. This is a relatively large cohort for a study of rare lung diseases and strengthens the findings. The international, multicenter design is especially important because ILD patients can differ by region, referral patterns, underlying causes, and clinical practice. A model that performs well across multiple centers is more likely to be useful in routine care.
Main Findings
The eNose performed well in distinguishing ILD subtypes. When different ILD groups were compared with all other ILDs combined, the area under the curve, or AUC, ranged from 0.88 to 0.92 in the training set and from 0.75 to 0.95 in the external validation set. In medical diagnostics, an AUC closer to 1.0 indicates better discrimination, so these values suggest strong performance overall.
When the investigators tried to separate the different ILD subtypes from each other, the AUC values ranged from 0.95 to 0.98 in the training set and from 0.83 to 0.93 in the validation set. These are promising results, showing that breath signatures may capture clinically meaningful differences between disease categories. Just as important, the model remained reasonably accurate when tested outside the original training centers, which supports its generalizability.
What These Results Mean
These findings suggest that eNose technology could help clinicians identify fibrotic interstitial lung disease more quickly and potentially distinguish one subtype from another without immediately resorting to invasive procedures. In practical terms, this may be valuable in clinics that manage complex respiratory patients, especially when diagnostic uncertainty is high.
For example, a noninvasive breath test might one day assist in differentiating idiopathic pulmonary fibrosis from other fibrotic ILDs, supporting earlier treatment decisions. It could also be used as an additional tool alongside imaging, lung function testing, serology, and clinical history rather than as a replacement for them.
Clinical Advantages of an Electronic Nose
An eNose has several appealing features. It is noninvasive, quick, and potentially easy to use at the point of care. Breath collection is generally simple and more comfortable for patients than biopsy-based testing. If integrated into clinical workflows, it could help triage patients, prioritize multidisciplinary review, and reduce diagnostic delay.
Another advantage is reproducibility across centers. The fact that this study included external validation suggests that the technology may be robust enough for broader clinical use, although further confirmation is still needed before routine adoption.
Important Limitations
Although the results are encouraging, this technology is not yet ready to replace current diagnostic standards. Breath profiles can be influenced by many factors, including smoking, diet, medications, infection, environmental exposures, and comorbid illnesses. Careful standardization of breath collection and data processing is essential.
Also, ILD diagnosis remains a complex clinical process. Imaging patterns, exposure history, autoimmune features, lung function trends, and multidisciplinary interpretation remain central to care. The eNose should be viewed as an adjunct tool that may improve diagnostic confidence, not as a stand-alone test.
Another limitation is that the study was conducted in expert centers, where patient selection and diagnostic accuracy may be higher than in general practice. Additional studies in broader populations, including earlier-stage disease and community settings, will be needed to determine how the test performs in everyday care.
Future Directions
The next step is likely to focus on prospective validation, standardization, and integration with other diagnostic tools. Researchers may explore whether combining eNose breath profiles with CT imaging, clinical variables, blood biomarkers, or machine learning improves accuracy further. Longitudinal studies could also determine whether breath signatures change with disease progression or treatment response.
If future studies confirm these findings, electronic nose technology could become part of a faster and less invasive diagnostic pathway for patients with suspected fibrotic ILD. That would be especially helpful for rare lung diseases, where time to diagnosis is often long and specialized expertise is not always immediately available.
Bottom Line
This international multicenter study shows that electronic nose technology can accurately distinguish breath profiles in patients with various fibrotic interstitial lung diseases. While it does not replace current diagnostic methods, it offers a promising noninvasive tool that may help clinicians identify and classify ILDs more efficiently. For patients, that could mean fewer invasive tests, faster answers, and earlier access to appropriate treatment.
