REPRODUCIBILITY OF TUMOR SEGMENTATION OUTCOMES WITH A DEEP LEARNING MODEL - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

REPRODUCIBILITY OF TUMOR SEGMENTATION OUTCOMES WITH A DEEP LEARNING MODEL

Résumé

In the last few years, there has been a growing awareness of reproducibility concerns in many areas of science. In this work, our goal is to evaluate the reproducibility of tumor segmentation outcomes produced with a deep segmentation model when MRI images are pre-processed (i) with two different versions of the same pre-processing pipeline, and (ii) by introducing numerical perturbations that mimic executions on different environments. Results show that these two variability sources can lead to important variations of segmentation outcomes: Dice can go as low as 0.59 and Hausdorff distance as high as 84.75. Moreover, both cases show a similar range of values, suggesting that the underlying causes for instability may be numerical stability. This work can be used as a benchmark to improve the numerical stability of the pipeline.
Fichier principal
Vignette du fichier
ISBI23_paper_final.pdf (861.53 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04006057 , version 1 (20-03-2023)

Identifiants

  • HAL Id : hal-04006057 , version 1

Citer

Morgane Des Ligneris, Axel Bonnet, Yohan Chatelain, Tristan Glatard, Michaël Sdika, et al.. REPRODUCIBILITY OF TUMOR SEGMENTATION OUTCOMES WITH A DEEP LEARNING MODEL. International Symposium on Biomedical Imaging (ISBI), Apr 2023, Cartagena de Indias, Colombia. ⟨hal-04006057⟩
150 Consultations
169 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More