BioSmart Inhale: AI-Enabled Breath Analysis for Early Asthma and Pulmonary Disorder Detection
Abstract Continuous monitoring and early detection of respiratory-related diseases, such as asthma and pulmonary disorders, are essential since these conditions may lead to life-threatening conditions. Despite the accuracy that conventional de- vices like gas chromatography-mass spectrometry (GC-MS) and infrared analyzer possess, these systems are usually expensive and bulky, thus limited to laboratory or clinical environments. To create an affordable and portable alter- native, the BioSmart Inhale project was initiated. BioSmart Inhale is a multi-sensor breath analysis system enhanced with artificial intelligence capabilities. Gas sensors are incorporated into this system for the identification of volatile organic compounds (VOCs). Apart from gas sensors, temperature and humidity sensors are included in the system for compensating the environmental impact and improving the performance. Compared to the current instruments, the system has incorporated a novel technology of biomarker detection based on the absorption of breath in this case. All acquired data will be analyzed with embedded machine learning algorithms, including SVM (Support Vector Ma- chine) and LSTM (Long Short-Term Memory). What makes this project unique is the fact that this instrument involves multi-sensors, artificial intelligence technologies, and the portability of the system. Thus, apart from monitoring, this device can be connected wirelessly and utilized in remote locations. BioSmart Inhale offers a practical and scalable solution for early disease detection, continuous monitoring, and personalized healthcare, making it suitable for both home and clinical use. Keywords: Breath analysis, Respiratory monitoring, Asthma detection, Biosensors, AI-based classification.
Authors
- Ushus S. Kumar (ORCID: https://orcid.org/0000-0002-8182-7012)
- R. Hariprasath
- S.K. Karventhan
- G. Sakthivel
Institutions
- SRM Institute of Science and Technology (IN)
- SRM Dental College (IN)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-04-30
- DOI
- https://doi.org/10.5281/zenodo.19909019
- Primary Topic
- Advanced Chemical Sensor Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00