Back

Machine Learning Analysis of Chest CT Scan Images as a Complementary Digital Test of Coronavirus (COVID-19) Patients

Al-karawi, D.; Al-Zaidi, S.; Polus, N.; Jassim, S.

2020-04-17 radiology and imaging
10.1101/2020.04.13.20063479 medRxiv
Show abstract

This paper reports on the development and performance of machine learning schemes for the analysis of Chest CT Scan images of Coronavirus COVID-19 patients and demonstrates significant success in efficiently and automatically testing for COVID-19 infection. In particular, an innovative frequency domain algorithm, to be called FFT-Gabor scheme, will be shown to predict in almost real-time the state of the patient with an average accuracy of 95.37%, sensitivity 95.99% and specificity 94.76%. The FFT-Gabor scheme is adequately informative in that clinicians can visually examine the FFT-Gabor feature to support their final diagnostic. Key StrengthsThe proposed FFT-Gabor scheme is an automatic machine learning scheme that works in real time and achieves significantly high accuracy with very low false negative, and can provide supporting evidences of the predicted decision by visually displaying the final features upon which decision is made. This scheme will be most beneficial when used in addition to the RT-PCR swab test of non-symptomatic cases.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.