Discriminative Hamiltonian variational autoencoder for accurate tumor segmentation in data-scarce regimes

Deep learning has gained significant attention in medical image segmentation. However, the limited availability of annotated training data presents a challenge to achieving accurate results. In efforts to overcome this challenge, data augmentation techniques have been proposed. However, the majority...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Neurocomputing (Amsterdam) Ročník 606; s. 128360
Hlavní autoři: Kebaili, Aghiles, Lapuyade-Lahorgue, Jérôme, Vera, Pierre, Ruan, Su
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 14.11.2024
Elsevier
Témata:
Data augmentation
Deep learning
Generative modeling
Hamiltonian Variational Autoencoder
Life Sciences
MRI
PET
Tumor segmentation
Variational Autoencoder
Deep learning
Generative modeling
Hamiltonian Variational Autoencoder
MRI
Variational Autoencoder
Data augmentation
Tumor segmentation
PET
Deep learning Data augmentation Tumor segmentation Generative modeling Variational Autoencoder MRI PET Hamiltonian Variational Autoencoder
ISSN:0925-2312
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Deep learning has gained significant attention in medical image segmentation. However, the limited availability of annotated training data presents a challenge to achieving accurate results. In efforts to overcome this challenge, data augmentation techniques have been proposed. However, the majority of these approaches primarily focus on image generation. For segmentation tasks, providing both images and their corresponding target masks is crucial, and the generation of diverse and realistic samples remains a complex task, especially when working with limited training datasets. To this end, we propose a new end-to-end hybrid architecture based on Hamiltonian Variational Autoencoders (HVAE) and a discriminative regularization to improve the quality of generated images. Our method provides an accurate estimation of the joint distribution of the images and masks, resulting in the generation of realistic medical images with reduced artifacts and off-distribution instances. As generating 3D volumes requires substantial time and memory, our architecture operates on a slice-by-slice basis to segment 3D volumes, capitalizing on the richly augmented dataset. Experiments conducted on two public datasets, BRATS (MRI modality) and HECKTOR (PET modality), demonstrate the efficacy of our proposed method on different medical imaging modalities with limited data.
ISSN:0925-2312
DOI:10.1016/j.neucom.2024.128360