Image Augmentation based on Variational Autoencoder for Breast Tumor Segmentation

Breast tumor segmentation based on Dynamic Contrast-Enhanced Magnetic Resonance Imaging is significant step for computable radiomics analysis of breast cancer. Manual tumor annotation is time-consuming process and involves medical acquaintance, biased, inclined to error, and inter-user discrepancy....

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Academic radiology Ročník 30 Suppl 2; s. S172
Hlavný autor: Balaji, K
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States 01.09.2023
Predmet:
Deep learning
Segmentation
Breast tumor
Dynamic contrast-enhanced magnetic resonance imaging
ISSN:1878-4046, 1878-4046
On-line prístup:Zistit podrobnosti o prístupe
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Breast tumor segmentation based on Dynamic Contrast-Enhanced Magnetic Resonance Imaging is significant step for computable radiomics analysis of breast cancer. Manual tumor annotation is time-consuming process and involves medical acquaintance, biased, inclined to error, and inter-user discrepancy. A number of modern trainings have revealed the capability of deep learning representations in image segmentation. Here, we describe a 3D Connected-UNets for tumor segmentation from 3D Magnetic Resonance Imagings based on encoder-decoder architecture. Due to a restricted training dataset size, a variational auto-encoder outlet is supplementary to renovate the input image itself in order to identify the shared decoder and execute additional controls on its layers. Based on initial segmentation of Connected-UNets, fully connected 3D provisional unsystematic domain is used to enhance segmentation outcomes by discovering 2D neighbor areas and 3D volume statistics. Moreover, 3D connected modules evaluation is used to endure around large modules and decrease segmentation noise. The proposed method has been assessed on two widely offered datasets, explicitly INbreast and the curated breast imaging subset of digital database for screening mammography The proposed model has also been estimated using a private dataset. The experimental results show that the proposed model outperforms the state-of-the-art methods for breast tumor segmentation.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1878-4046
1878-4046
DOI:10.1016/j.acra.2022.12.035