@article{MacLean_Ebadi_Florea_XI_Wong_2022, title={An Initial Study into the Feasibility of Deep Learning-Based COVID-19 Severity Classification using Point-of-Care Ultrasound Imaging}, volume={7}, url={https://openjournals.uwaterloo.ca/index.php/vsl/article/view/4902}, DOI={10.15353/jcvis.v7i1.4902}, abstractNote={<p>Integral to the treatment of patients suffering from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is assessment of the severity of the illness, allowing clinicians to more effectively apply care and devise a plan of treatment. Since the workload of clinicians is high at the best of times, let alone during a global pandemic, much work has gone into creating computer-aided clinical decision support systems, often enabled by deep learning tools. Previous work has investigated the ability to identify COVID-19 positive patients from point-of-care ultrasound (POCUS) images, but decision support systems for POCUS-based COVID-19 severity stratification have not yet been presented. In this study, we examine the feasibility of using a deep learning neural network architecture to classify POCUS images from an open source repository into distinct severity levels based on annotations from an experienced doctor of emergency medicine, hopefully leading to the implementation of such a system into a real-world clinical workflow.</p>}, number={1}, journal={Journal of Computational Vision and Imaging Systems}, author={MacLean, Alexander and Ebadi, Ashkan and Florea, Adrian and XI, PENGCHENG and Wong, Alexander}, year={2022}, month={Apr.}, pages={31–33} }