In this section
@article{Davies:2024:10.1109/OJEMB.2024.3360688,
author = {Davies, HJ and Hammour, G and Xiao, H and Bachtiger, P and Larionov, A and Molyneaux, PL and Peters, NS and Mandic, DP},
doi = {10.1109/OJEMB.2024.3360688},
journal = {IEEE Open J Eng Med Biol},
pages = {148--156},
title = {Physically meaningful surrogate data for COPD},
url = {http://dx.doi.org/10.1109/OJEMB.2024.3360688},
volume = {5},
year = {2024}
}
TY - JOUR
AB - The rapidly increasing prevalence of debilitating breathing disorders, such as chronic obstructive pulmonary disease (COPD), calls for a meaningful integration of artificial intelligence (AI) into respiratory healthcare. Deep learning techniques are "data hungry" whilst patient-based data is invariably expensive and time consuming to record. To this end, we introduce a novel COPD-simulator, a physical apparatus with an easy to replicate design which enables rapid and effective generation of a wide range of COPD-like data from healthy subjects, for enhanced training of deep learning frameworks. To ensure the faithfulness of our domain-aware COPD surrogates, the generated waveforms are examined through both flow waveforms and photoplethysmography (PPG) waveforms (as a proxy for intrathoracic pressure) in terms of duty cycle, sample entropy, FEV1/FVC ratios and flow-volume loops. The proposed simulator operates on healthy subjects and is able to generate FEV1/FVC obstruction ratios ranging from greater than 0.8 to less than 0.2, mirroring values that can observed in the full spectrum of real-world COPD. As a final stage of verification, a simple convolutional neural network is trained on surrogate data alone, and is used to accurately detect COPD in real-world patients. When training solely on surrogate data, and testing on real-world data, a comparison of true positive rate against false positive rate yields an area under the curve of 0.75, compared with 0.63 when training solely on real-world data.
AU - Davies,HJ
AU - Hammour,G
AU - Xiao,H
AU - Bachtiger,P
AU - Larionov,A
AU - Molyneaux,PL
AU - Peters,NS
AU - Mandic,DP
DO - 10.1109/OJEMB.2024.3360688
EP - 156
PY - 2024///
SN - 2644-1276
SP - 148
TI - Physically meaningful surrogate data for COPD
T2 - IEEE Open J Eng Med Biol
UR - http://dx.doi.org/10.1109/OJEMB.2024.3360688
UR - https://www.ncbi.nlm.nih.gov/pubmed/38487098
UR - https://ieeexplore.ieee.org/document/10417113
UR - http://hdl.handle.net/10044/1/112188
VL - 5
ER -
For any enquiries about the Fungal Science Network at Imperial, please contact: