Two-Stage COVID19 Classification Using BERT Features

TL;DR

This paper introduces a two-stage COVID-19 diagnosis method from lung CT scans using a novel combination of 3D-CNN and BERT for feature extraction and aggregation, achieving high accuracy.

Contribution

It presents a novel two-stage framework combining 3D-CNN and BERT for effective COVID-19 classification from CT scans, improving accuracy over existing methods.

Findings

Achieved macro F1 score of 0.9164 on validation data.

Demonstrated effective aggregation of features using BERT.

Improved COVID-19 detection accuracy from lung CT scans.

Abstract

We propose an automatic COVID1-19 diagnosis framework from lung CT-scan slice images using double BERT feature extraction. In the first BERT feature extraction, A 3D-CNN is first used to extract CNN internal feature maps. Instead of using the global average pooling, a late BERT temporal pooing is used to aggregate the temporal information in these feature maps, followed by a classification layer. This 3D-CNN-BERT classification network is first trained on sampled fixed number of slice images from every original CT scan volume. In the second stage, the 3D-CNN-BERT embedding features are extracted on all slice images of every CT scan volume, and these features are averaged into a fixed number of segments. Then another BERT network is used to aggregate these multiple features into a single feature followed by another classification layer. The classification results of both stages are combined to generate final outputs. On the validation dataset, we achieve macro F1 score of 0.9164.