Reproduce “Patient-specific modelling, simulation and real-time processing for respiratory diseases” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Stavros Nousias
This dissertation develops patient-specific computational models of the bronchial tree for respiratory diseases like asthma, enabling simulation of airflow dynamics in normal and constricted lungs, and introduces real-time, computationally efficient audio classification methods for monitoring medication adherence with pressurized metered-dose inhalers. It presents the AVATREE framework for anatomical modeling and demonstrates a CNN-based audio classification approach achieving 94.12% accuracy in mixed, unseen patient scenarios while being 15-40 times faster than other methods.
Key insights
- Patient-specific digital twins, generated by combining CT scans with advanced space-filling and geometry processing algorithms, can accurately represent complex pulmonary structures and simulate disease states like bronchoconstriction.
- CFD simulations on these patient-specific models reveal quantitative changes in airflow dynamics and potential drug deposition patterns, highlighting how individual anatomy and disease impact therapeutic delivery.
- A CNN-based approach for audio event classification from inhaler use can provide highly accurate and computationally inexpensive real-time monitoring, suitable for embedded systems, even in challenging mixed audio environments and for unseen subjects.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.