CoRSAI: A System for Robust Interpretation of CT Scans of COVID-19 Patients Using Deep Learning

TL;DR

This paper introduces CoRSAI, a deep learning system that accurately segments lung lesions in COVID-19 CT scans, estimates affected lung volume, and assesses disease severity, outperforming radiologists.

Contribution

The novel ensemble deep learning approach improves segmentation accuracy and severity classification of COVID-19 lung damage over expert radiologists.

Findings

Model outperforms most radiologists in segmentation.

Model surpasses all radiologists in severity classification.

Effective multi-center training enhances robustness.

Abstract

Analysis of chest CT scans can be used in detecting parts of lungs that are affected by infectious diseases such as COVID-19.Determining the volume of lungs affected by lesions is essential for formulating treatment recommendations and prioritizingpatients by severity of the disease. In this paper we adopted an approach based on using an ensemble of deep convolutionalneural networks for segmentation of slices of lung CT scans. Using our models we are able to segment the lesions, evaluatepatients dynamics, estimate relative volume of lungs affected by lesions and evaluate the lung damage stage. Our modelswere trained on data from different medical centers. We compared predictions of our models with those of six experiencedradiologists and our segmentation model outperformed most of them. On the task of classification of disease severity, ourmodel outperformed all the radiologists.