Projected pooling loss for red nucleus segmentation with soft topology constraints - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Medical Imaging Année : 2024

Projected pooling loss for red nucleus segmentation with soft topology constraints

Résumé

Purpose. Deep learning is the standard for medical image segmentation. However, it may encounter difficulties when the training set is small. Also, it may generate anatomically aberrant segmentations. Anatomical knowledge can be potentially useful as a constraint in deep learning segmentation methods. In this paper, we propose a novel loss function based on projected pooling to introduce soft topological contraints. Our main application is the segmentation of the red nucleus from quantitative susceptibility mapping (QSM) which is of interest in parkinsonian syndromes. Approach. This new loss function introduces soft constraints on the topology by magnifying small parts of the structure to segment to avoid that they are discarded in the segmentation process. To that purpose, we use projection of the structure onto the three planes and then use a series of MaxPooling operations with increasing kernel sizes. These operations are performed both for the ground-truth and the prediction and the difference is computed to obtain the loss function. As a result, it can reduce topological errors as well as defects in the structure boundary. The approach is easy to implement and computationally efficient. Results. When applied to the segmentation of the red nucleus from QSM data, the approach led to a very high accuracy (Dice 89.9%) and no topological errors. Moreover, the proposed loss function improved the Dice accuracy over the baseline when the training set was small. We also studied three tasks from the medical segmentation decathlon challenge (MSD) (heart, spleen, and hippocampus). For the MSD tasks, the Dice accuracies were similar for both approaches but the topological errors were reduced. Conclusions. We proposed an effective method to automatically segment the red nucleus which is based on a new loss for introducing topology constraints in deep learning segmentation.
Fichier principal
Vignette du fichier
Guanghui_topological_loss_journal (2).pdf (12.16 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04643094 , version 1 (10-07-2024)

Identifiants

Citer

Guanghui Fu, Rosana El Jurdi, Lydia Chougar, Didier Dormont, Romain Valabregue, et al.. Projected pooling loss for red nucleus segmentation with soft topology constraints. Journal of Medical Imaging, 2024, 11 (04), ⟨10.1117/1.JMI.11.4.044002⟩. ⟨hal-04643094⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More