BreathBase® Platform

Immediate feedback.

From high-quality breath measurements to the generation and validation of diagnostic models and composite biomarkers.


The innovative idea of the BreathBase Platform was conceived by Breathomix and developed in collaboration with our technology partners Microsoft and Tecknoworks.

BreathBase is an IoT based solution that works via the SpiroNose, which is connected to an internet enabled IoT device (Gateway) that real-time routes the measurement data to a processing component, necessary for analysis and interpretation. The results are presented in the BreathBase Web Application. BreathBase is uniquely designed to allow real-time analysis of the sensor data based on advanced signal processing and artificial intelligence (AI), providing diagnostic feedback to the user within seconds. The BreathBase Platform is developed conform the requirements of the following standard, ISO 27001 (Information Security), and is hosted on the Microsoft Azure Western Europe data centre in Middenmeer the Netherlands. Together with Microsoft and Tecknoworks, Breathomix developed the architecture details for implementing this solution on the Azure Cloud with the necessary scalability, security and availability in mind. This makes BreathBase an easy-to-use, turn-key solution to any clinical or academic partner in the world.

All-in-one Solution

We offer a complete set of hardware and software for the analysis of non-invasive biomarkers in exhaled breath. From high quality measurement of VOCs to the generation and validation of diagnostic models using our online analytical platform. With an integrated reference breath database and high quality standards our platform provides reliable, reproducible and immediate results.

Benefits of BreathBase® over other sensing technologies

Broadly speaking, there are two approaches to exhaled breath analysis. Methods based on mass spectrometry (MS) aim to detect individual VOCs and in most cases identify them, which is particularly useful for pathophysiologic research. Even though this represents the standard for analytical chemistry, these methods are of limited use in clinical settings, because they are time consuming and require offline laboratory procedures using expensive equipment run by highly trained personnel.

On the other hand, eNoses are based on cross-reactive nonspecific sensor arrays purposely not identifying individual VOCs. The VOCs competitively interact with the sensors allowing multiple VOCs to bind to the same sensor based on their affinity for both the sensor and its substrate. Likewise, multiple sensors interact with the same volatile. Notably, this is comparable to the powerful mammalian olfactory system and results in a pattern of firing sensors that is driven by the complete mixture of VOCs. Thereby, cutting-edge pattern recognition by eNoses allows probabilistic analyses suitable for clinical diagnostics and precision medicine at the point-of-care.

The inability to identify individual VOCs does not impede the clinical application of eNose technology in any way. However, unlike BreathBase, thus far many eNoses suffer from stability, reproducibility and accuracy issues such as poor intra-device repeatability and limited temporal stability. These problems have hampered the translation of eNose technology from being a research tool to a point-of-care test.

BreathBase overcomes these problems. The sensor data obtained by the SpiroNose is reproducible and interchangeable, and the BreathBase analysis algorithms have shown adequate transferability between devices (De Vries ERJ 2018 & De Vries JBR 2015). The latter allows for the exchange of results between different labs and locations, which is a prerequisite for clinical implementation. Additionally, the powerful pattern recognition algorithms of BreathBase ignore chemical noise and are able to detect the VOC patterns of interest with high accuracy values. Finally, BreathBase allows for real-time and online analysis of the SpiroNose sensor data and provides immediate feedback at the point-of-care.