A Computationally-Inexpensive Strategy in CT Image Data Augmentation for
Robust Deep Learning Classification of COVID-19
Abstract
Coronavirus disease 2019 (COVID-19) has spread globally for two years,
and chest computed tomography (CT) has been used to diagnose COVID-19
and identify lung damage in long COVID-19 patients. At the beginning of
the epidemic, there was a shortage of large and publicly available CT
datasets due to privacy concerns. Therefore, it is important to classify
CT scans correctly when only limited resources are available, as it will
happen again in future pandemics. We followed the transfer learning
procedure and limited hyperparameters to use as few computing resources
as possible. The Advanced Normalisation Tools (ANTs) were used to
synthesise images as augmented/independent data and trained on
EfficientNet to investigate the effect of synthetic images. On the
COVID-CT dataset, classification accuracy increased from 91.15% to
95.50% and Area Under the Receiver Operating Characteristic (AUC) from
96.40% to 98.54%. We also customised a small dataset to simulate data
collected in the early stages of the outbreak and improve accuracy from
85.95% to 94.32% and AUC from 93.21% to 98.61%. This paper provides
a feasible solution with a relatively low computational cost for medical
image classification when scarce data are available and traditional data
augmentation may fail.
This work has been submitted to the IEEE for possible publication.
Copyright may be transferred without notice, after which this version
may no longer be accessible.