Deep Learning Assisted Covid-19 Detection using full CT-scans
preprintposted on 30.10.2020, 18:31 by varan singhrohila, Nitin Gupta, Amit Kaul, Deepak Sharma
The ongoing pandemic of COVID-19 has shown
the limitations of our current medical institutions. There
is a need for research in the field of automated diagnosis
for speeding up the process while maintaining accuracy
and reducing computational requirements. In this work, an
automatic diagnosis of COVID-19 infection from CT scans
of the patients using Deep Learning technique is proposed.
The proposed model, ReCOV-101 uses full chest CT scans to
detect varying degrees of COVID-19 infection, and requires
less computational power. Moreover, in order to improve
the detection accuracy the CT-scans were preprocessed by
employing segmentation and interpolation. The proposed
scheme is based on the residual network, taking advantage
of skip connection, allowing the model to go deeper.
Moreover, the model was trained on a single enterpriselevel
GPU such that it can easily be provided on the edge of
the network, reducing communication with the cloud often
required for processing the data. The objective of this work
is to demonstrate a less hardware-intensive approach for COVID-19 detection with excellent performance that can
be combined with medical equipment and help ease the
examination procedure. Moreover, with the proposed model
an accuracy of 94.9% was achieved.