AI-powered CPET analysis
Oxynet
Where AI meets CPET. An open toolset for the automatic interpretation of cardiopulmonary exercise test data — built with deep learning.
The Project
About Oxynet
Universal access to high-quality healthcare remains a global challenge. Oxynet leverages AI and vast data resources to revolutionize the diagnosis of medical conditions through CPET analysis, enabling accurate and timely clinical decisions.
CPET Experts
A global network of exercise physiologists and clinicians providing labeled training data, clinical validation, and domain expertise.
Crowdsourced Dataset
A large, continuously growing dataset collected across diverse clinical settings worldwide, enabling robust and generalizable model training.
Advanced AI
Deep neural networks built with Keras and TensorFlow that approximate expert human judgment in CPET interpretation with high accuracy.
We actively seek collaboration with universities, hospitals, clinics, medical professionals, and companies. Together we can advance research, contribute to publications, share data, and validate algorithms for clinical implementation.
Get in touch →See it in action
From Raw Data to Clinical Insight
Drag the slider to see how Oxynet transforms raw CPET measurements into intensity domains — automatically detecting LT and RCP to classify every breath.
Intensity Domains
Moderate Domain
Below LTVO₂ reaches steady state within minutes. Blood lactate returns to resting levels. Exercise is fully sustainable.
Heavy Domain
LT → RCPA VO₂ slow component emerges. Lactate rises but stabilises above baseline. Prolonged exercise remains possible.
Severe Domain
Above RCPRespiratory compensation is engaged. Lactate and VO₂ rise continuously toward VO₂max. Exercise tolerance is time-limited.
Based on Keir et al., Sports Medicine (2022)
Open Source
The Pyoxynet Package
A comprehensive suite of deep neural network algorithms specifically designed for CPET data analysis. Built with Keras and TensorFlow, models available in efficient TFLite format.
Inference Model
Estimates exercise intensity domains from CPET data with high accuracy. Supports VO₂, VCO₂, VE, PetO₂, PetCO₂, VE/VO₂, and VE/VCO₂ inputs.
Generator Model
Creates realistic synthetic CPET data for research and validation using a Conditional GAN (CGAN) architecture.
Quick Start
Code Examples
Requires Python 3.8+. Pyoxynet automatically handles data interpolation and supports second-by-second, breath-by-breath, and averaged CPET data formats.
pip install pyoxynetResearch
Scientific Publications
Peer-reviewed research, reviews, and articles behind the Oxynet project.
Get in Touch
Contact
Interested in collaboration or have questions about the project?
Feedback & Issues
Oxynet Team
Bug reports, feature requests, and general inquiries about the Oxynet project.
oxynetcpetinterpreter@gmail.comPrincipal Investigator
Andrea Zignoli
Research collaborations, academic partnerships, and scientific enquiries.
andrea.zignoli@unitn.it