Evaluating Volumetric and Slice-Based Approaches for COVID-19 Detection in Chest CTs

Abstract

The paper presents a comparative analysis of several distinct approaches based on deep learning for identifying COVID-19 cases in chest CTs. A first approach is a volumetric one, involving 3D convolutions, while other two approaches perform at first slice-wise classification and then aggregate the results at the volume level. The experiments are carried on the COV19-CT-DB dataset, with the aim of addressing the challenge raised by the MIA-COV19D Competition within ICCV 2021. Our best results reach a macro F1 score of 92.34% on the validation subset and 90.06% on the test set, obtained with the volumetric approach which was ranked second in the competition. Its performance can be further improved by a simple trick, using semi-supervised training in the form of self-training, technique which proved to bring a consistent increase over the reported F1-score on the validation subset.

Cite

Text

Miron et al. "Evaluating Volumetric and Slice-Based Approaches for COVID-19 Detection in Chest CTs." IEEE/CVF International Conference on Computer Vision Workshops, 2021. doi:10.1109/ICCVW54120.2021.00065

Markdown

[Miron et al. "Evaluating Volumetric and Slice-Based Approaches for COVID-19 Detection in Chest CTs." IEEE/CVF International Conference on Computer Vision Workshops, 2021.](https://mlanthology.org/iccvw/2021/miron2021iccvw-evaluating/) doi:10.1109/ICCVW54120.2021.00065

BibTeX

@inproceedings{miron2021iccvw-evaluating,
  title     = {{Evaluating Volumetric and Slice-Based Approaches for COVID-19 Detection in Chest CTs}},
  author    = {Miron, Radu and Moisii, Cosmin and Dinu, Sergiu and Breaban, Mihaela Elena},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2021},
  pages     = {529-536},
  doi       = {10.1109/ICCVW54120.2021.00065},
  url       = {https://mlanthology.org/iccvw/2021/miron2021iccvw-evaluating/}
}