Artificial Intelligence-Powered Nanosensor Platforms for Non-Invasive Breathomic Diagnostics
Issued Date
2025-01-01
Resource Type
eISSN
11778903
Scopus ID
2-s2.0-105026444918
Journal Title
Nanotechnology Science and Applications
Volume
18
Start Page
611
End Page
641
Rights Holder(s)
SCOPUS
Bibliographic Citation
Nanotechnology Science and Applications Vol.18 (2025) , 611-641
Suggested Citation
Chaudhary V., Bhadola P. Artificial Intelligence-Powered Nanosensor Platforms for Non-Invasive Breathomic Diagnostics. Nanotechnology Science and Applications Vol.18 (2025) , 611-641. 641. doi:10.2147/NSA.S546714 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/113980
Title
Artificial Intelligence-Powered Nanosensor Platforms for Non-Invasive Breathomic Diagnostics
Author(s)
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Global healthcare settings are increasingly burdened by critical diseases, where conventional diagnostics are often expensive, invasive, time-consuming and centralised. It creates a critical gap for rapid, accessible, portable and non-invasive health assessment. AI-powered Nanosensors for Breathomics Diagnostics (AND) platforms have emerged as a transformative solution to this complex global problem, integrating highly sensitive nanomaterials with advanced machine intelligence to detect disease biomarkers in exhaled breath. These platforms have already demonstrated high performance, with reports of 90–95% diagnostic accuracy for conditions such as lung cancer and achieving sub-ppb detection limits. These platforms are not limited to controlled laboratory settings but have been employed to monitor a spectrum of diseases, including cancer, asthma, diabetes, coronavirus disease, and renal failure. Their integration into wearable systems, smartphones, smart masks and multimodal laboratory systems further extends their applications in predictive analytics, personalised medicine and real-time human–machine interaction. However, challenges related to data standardisation, sensor selectivity, ethical AI, and clinical validation have limited their commercialization. It necessitates solutions such as Explainable AI, physics-informed modelling, network theory, and the development of large-scale clinical breath databases to enhance clinical reliability, model robustness, diagnose sensor drift, and attain transparency. This article critically details the recent progress and charts a new path forward for translating AND platforms from research to clinical reality as next-generation healthcare.