MedmeshCNN - Enabling meshcnn for medical surface models.
Convolutional neural network
Geometric deep learning
Intracranial aneurysms
Mesh processing
Shape segmentation
Surface models
Journal
Computer methods and programs in biomedicine
ISSN: 1872-7565
Titre abrégé: Comput Methods Programs Biomed
Pays: Ireland
ID NLM: 8506513
Informations de publication
Date de publication:
Oct 2021
Oct 2021
Historique:
received:
23
09
2020
accepted:
20
08
2021
pubmed:
3
9
2021
medline:
29
9
2021
entrez:
2
9
2021
Statut:
ppublish
Résumé
MeshCNN is a recently proposed Deep Learning framework that drew attention due to its direct operation on irregular, non-uniform 3D meshes. It outperformed state-of-the-art methods in classification and segmentation tasks of popular benchmarking datasets. The medical domain provides a large amount of complex 3D surface models that may benefit from processing with MeshCNN. However, several limitations prevent outstanding performances on highly diverse medical surface models. Within this work, we propose MedMeshCNN as an expansion dedicated to complex, diverse, and fine-grained medical data. MedMeshCNN follows the functionality of MeshCNN with a significantly increased memory efficiency that allows retaining patient-specific properties during processing. Furthermore, it enables the segmentation of pathological structures that often come with highly imbalanced class distributions. MedMeshCNN achieved an Intersection over Union of 63.24% on a highly complex part segmentation task of intracranial aneurysms and their surrounding vessel structures. Pathological aneurysms were segmented with an Intersection over Union of 71.4%. MedMeshCNN enables the application of MeshCNN on complex, fine-grained medical surface meshes. It considers imbalanced class distributions derived from pathological findings and retains patient-specific properties during processing.
Sections du résumé
BACKGROUND AND OBJECTIVE
OBJECTIVE
MeshCNN is a recently proposed Deep Learning framework that drew attention due to its direct operation on irregular, non-uniform 3D meshes. It outperformed state-of-the-art methods in classification and segmentation tasks of popular benchmarking datasets. The medical domain provides a large amount of complex 3D surface models that may benefit from processing with MeshCNN. However, several limitations prevent outstanding performances on highly diverse medical surface models. Within this work, we propose MedMeshCNN as an expansion dedicated to complex, diverse, and fine-grained medical data.
METHODS
METHODS
MedMeshCNN follows the functionality of MeshCNN with a significantly increased memory efficiency that allows retaining patient-specific properties during processing. Furthermore, it enables the segmentation of pathological structures that often come with highly imbalanced class distributions.
RESULTS
RESULTS
MedMeshCNN achieved an Intersection over Union of 63.24% on a highly complex part segmentation task of intracranial aneurysms and their surrounding vessel structures. Pathological aneurysms were segmented with an Intersection over Union of 71.4%.
CONCLUSIONS
CONCLUSIONS
MedMeshCNN enables the application of MeshCNN on complex, fine-grained medical surface meshes. It considers imbalanced class distributions derived from pathological findings and retains patient-specific properties during processing.
Identifiants
pubmed: 34474194
pii: S0169-2607(21)00446-6
doi: 10.1016/j.cmpb.2021.106372
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
106372Informations de copyright
Copyright © 2021 Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors Lisa Schneider, Annika Niemann, Oliver Beuing, Bernhard Preim, Sylvia Saalfeld declare that there is no conflict of interest