Approaching the Semantic Segmentation in Medical Problems: A Solution for Pneumothorax Detection

Călin Timbus, Vlad Miclea, Camelia Lemnaru

2021

Abstract

We present a method for detecting and delineating pneumothorax from X-Ray medical images by using a threestep processing pipeline: a deep learning classification module, responsible for detecting the possible existence of a collapsed lung within an image, followed by a segmentation model applied on the positive samples (as detected by the classification module). The last module attempts to eliminate possible artefacts based on their size. We demonstrate how the pipeline employed significantly improves the results, by increasing the mean-Dice coefficient metric by 0.13, in comparison with the performance of a single segmentation module. In addition to this, we demonstrate that using together specific state-of-the-art techniques leads to improved results, without employing techniques such as dataset enrichment from external sources, semi-supervised learning or pretraining on much larger medical datasets.

Download


Paper Citation


in Harvard Style

Timbus C., Miclea V. and Lemnaru C. (2021). Approaching the Semantic Segmentation in Medical Problems: A Solution for Pneumothorax Detection. In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP; ISBN 978-989-758-488-6, SciTePress, pages 265-272. DOI: 10.5220/0010185402650272


in Bibtex Style

@conference{visapp21,
author={Călin Timbus and Vlad Miclea and Camelia Lemnaru},
title={Approaching the Semantic Segmentation in Medical Problems: A Solution for Pneumothorax Detection},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP},
year={2021},
pages={265-272},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010185402650272},
isbn={978-989-758-488-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP
TI - Approaching the Semantic Segmentation in Medical Problems: A Solution for Pneumothorax Detection
SN - 978-989-758-488-6
AU - Timbus C.
AU - Miclea V.
AU - Lemnaru C.
PY - 2021
SP - 265
EP - 272
DO - 10.5220/0010185402650272
PB - SciTePress