Detail publikačního výsledku

Segmentation of spinal cord structures from MRI images using deep learning

KREJČÍ, K.; CHMELÍK, J.; VALOSEK, J.

Originální název

Segmentation of spinal cord structures from MRI images using deep learning

Anglický název

Segmentation of spinal cord structures from MRI images using deep learning

Druh

Stať ve sborníku mimo WoS a Scopus

Originální abstrakt

Morphometric measures obtained from magnetic resonance images (MRI) are commonly used to assess the severity of spinal cord compression. However, current automatic methods do not perform sufficiently when applied to T1-weighted MRI images. The purpose of this work is to develop automatic deep learning models for spinal cord and gray matter segmentation from T1-weighted images. The proposed models were trained on a dataset of 66 healthy subjects and compared with state-of- the-art methods. The preliminary results demonstrated a higher Dice coefficient and lower Hausdorff distance compared to the existing methods suggesting promising model performance.

Anglický abstrakt

Morphometric measures obtained from magnetic resonance images (MRI) are commonly used to assess the severity of spinal cord compression. However, current automatic methods do not perform sufficiently when applied to T1-weighted MRI images. The purpose of this work is to develop automatic deep learning models for spinal cord and gray matter segmentation from T1-weighted images. The proposed models were trained on a dataset of 66 healthy subjects and compared with state-of- the-art methods. The preliminary results demonstrated a higher Dice coefficient and lower Hausdorff distance compared to the existing methods suggesting promising model performance.

Klíčová slova

Magnetic Resonance Imaging, Deep Learning, Segmentation, Spinal Cord, Spinal Cord Compression

Klíčová slova v angličtině

Magnetic Resonance Imaging, Deep Learning, Segmentation, Spinal Cord, Spinal Cord Compression

Autoři

KREJČÍ, K.; CHMELÍK, J.; VALOSEK, J.

Vydáno

25.04.2023

Nakladatel

Brno University of Technology, Faculty of Electrical Engineering and Communication

Místo

Brno, Czech Republic

ISBN

978-80-214-6153-6

Kniha

Proceedings I of the 29th Conference STUDENT EEICT 2023

Strany od

52

Strany do

55

Strany počet

4

URL

https://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_1.pdf

BibTex

@inproceedings{BUT200196,
  author="Kateřina {Krejčí} and Jiří {Chmelík} and  {}",
  title="Segmentation of spinal cord structures from MRI
images using deep learning",
  booktitle="Proceedings I of the 29th Conference STUDENT EEICT 2023",
  year="2023",
  pages="52--55",
  publisher="Brno University of Technology, Faculty of Electrical Engineering and Communication",
  address="Brno, Czech Republic",
  isbn="978-80-214-6153-6",
  url="https://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_1.pdf"
}