Detail publikace

Deep Learning Pipeline for Chromosome Segmentation

PIJÁČKOVÁ, K. GÖTTHANS, T. GÖTTHANS, J.

Originální název

Deep Learning Pipeline for Chromosome Segmentation

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

angličtina

Originální abstrakt

Chromosome segmentation is a challenging and time-consuming part of karyotyping and requires a high level of expertise. Computer segmentation algorithms still require the assistance of cytologists in more complicated cases with overlapping or touching chromosomes. Deep learning models have the potential to make the segmentation process completely automated, and their applications are currently actively re-searched. This paper proposes a segmentation pipeline by using deep learning models and traditional computer vision algorithms. This process can be split into four steps, in which we use U-Net architecture to remove any background noises of the metaphase image. Next, we use thresholding and skeletonization to extract and classify single chromosomes and chromosome clusters. As a final step, we use Mask R-CNN, for instance, segmentation on the overlapping and touching chromosomes, and apply test-time augmentation to improve the model's precision.

Klíčová slova

chromosome segmentation, karyotyping, deep learning, image processing, instance segmentation, test-time augmentation

Autoři

PIJÁČKOVÁ, K.; GÖTTHANS, T.; GÖTTHANS, J.

Vydáno

3. 5. 2022

Nakladatel

IEEE

ISBN

978-1-7281-8686-3

Kniha

2022 32nd International Conference Radioelektronika (RADIOELEKTRONIKA)

Strany od

197

Strany do

201

Strany počet

5

URL

https://ieeexplore.ieee.org/document/9764950

BibTex

@inproceedings{BUT178914,
  author="Kristýna {Pijáčková} and Tomáš {Götthans} and Jakub {Götthans}",
  title="Deep Learning Pipeline for Chromosome Segmentation",
  booktitle="2022 32nd International Conference Radioelektronika (RADIOELEKTRONIKA)",
  year="2022",
  pages="197--201",
  publisher="IEEE",
  doi="10.1109/RADIOELEKTRONIKA54537.2022.9764950",
  isbn="978-1-7281-8686-3",
  url="https://ieeexplore.ieee.org/document/9764950"
}