Oscillometry and Artificial Intelligence (AI) Detected Adventitious Lung Sounds During Exercise Induced Bronchoconstriction in Children
Journal of Pediatric Advance Research · 17.05.2026 · 5 (2) · 1
DOI: 10.46889/jpar.2026.5204 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Pediatric Advance Research, online erschienen am 17.05.2026. Verfasst von Mario Barreto und 8 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Künstliche Intelligenz in der Pflege, Kinderkrankenpflege
Abstract (Originalsprache)
Background
Exercise Induced Bronchoconstriction (EIB) is a common cause of exercise limitation in children, yet adventitious lung sounds after exercise are often scarce and poorly correlated with spirometric impairment. Artificial Intelligence (AI) assisted electronic stethoscopes may improve sound detection, while oscillometry provides sensitive measures of small airway mechanics. Whether post exercise adventitious sounds relate to oscillometric changes in pediatric patients remains unclear.
Objectives
To evaluate whether AI detected adventitious lung sounds after exercise reflect oscillometric or spirometric changes in children with exercise induced symptoms and to assess agreement between AI assisted and physician auscultation.
Methods
Children with recurrent exercise induced symptoms underwent baseline assessments, treadmill exercise testing and post exercise spirometry, oscillometry and AI assisted lung sound recordings (StethoMe®). Adventitious sounds (wheezes/rhonchi) were compared with lung function outcomes and with physician auscultation. EIB was defined as a ≥10% fall in FEV₁. Associations between sound intensity and oscillometric parameters were evaluated.
Results
Eighty nine children (mean age 11.6 ± 2.6 years) completed the protocol; 21 (23.6%) had EIB. Adventitious sounds were infrequent overall but more common after exercise. Agreement between the AI stethoscope and the physician was substantial for rhonchi (κ = 0.77) and excellent for wheezes (κ = 0.92) post exercise. Children with any post exercise adventitious sound (n = 18) showed higher expiratory R (R5, Fdep) and markedly lower expiratory X across all frequencies compared with those without sounds, whereas spirometric indices did not differ. Sound intensity correlated with post exercise expiratory R5, Fdep and especially expiratory X5. Inspiratory oscillometric values showed no meaningful associations.
Conclusion
AI detected adventitious lung sounds after exercise reflect expiratory oscillometric abnormalities but are not linked to spirometry defined EIB. Expiratory reactance, in particular, mirrors the presence and intensity of wheezes and rhonchi. Integrating AI assisted auscultation with oscillometry may provide complementary insights into small airway mechanics during exercise induced airway narrowing.
Zitieren
APA
Barreto, M., Tripodi, S., Fracasso, M., Cerocchi, M., Mazzuca, M. C., Evangelisti, M., Pagani, J., Guglielmi, F., & Parisi, P. (2026). Oscillometry and Artificial Intelligence (AI) Detected Adventitious Lung Sounds During Exercise Induced Bronchoconstriction in Children. Journal of Pediatric Advance Research, 5(2), 1. https://doi.org/10.46889/jpar.2026.5204
Vancouver
Barreto M, Tripodi S, Fracasso M, Cerocchi M, Mazzuca MC, Evangelisti M, et al. Oscillometry and Artificial Intelligence (AI) Detected Adventitious Lung Sounds During Exercise Induced Bronchoconstriction in Children. Journal of Pediatric Advance Research. 2026;5(2):1. doi:10.46889/jpar.2026.5204
BibTeX
@article{Barreto2026,
author = {Barreto, Mario and Tripodi, Salvatore and Fracasso, Matteo and Cerocchi, Martina and Mazzuca, Maria Cristina and Evangelisti, Melania and Pagani, Jacopo and Guglielmi, Francesco and Parisi, Pasquale},
title = {Oscillometry and Artificial Intelligence (AI) Detected Adventitious Lung Sounds During Exercise Induced Bronchoconstriction in Children},
journal = {Journal of Pediatric Advance Research},
year = {2026},
volume = {5},
number = {2},
pages = {1},
doi = {10.46889/jpar.2026.5204}
}
EUVIMED