Image reconstruction from projections of digital breast tomosynthesis using deep learning
Carregando...
Fontes externas
Fontes externas
Data
Orientador
Coorientador
Pós-graduação
Curso de graduação
Título da Revista
ISSN da Revista
Título de Volume
Editor
SPIE, the international society for optics and photonics
Tipo
Trabalho apresentado em evento
Direito de acesso
Acesso restrito
Fontes externas
Fontes externas
Resumo
The Filtered Backprojection (FBP) algorithm for Computed Tomography (CT) reconstruction can be mapped entire in an Artificial Neural Network (ANN), with the backprojection (BP) operation simulated analytically in a layer and the Ram-Lak filter simulated as a convolutional layer. Thus, this work adapts the BP layer for Digital Breast Tomosynthesis (DBT) reconstruction, making possible the use of FBP simulated as an ANN to reconstruct DBT images. We showed that making the Ram-Lak layer trainable, the reconstructed image can be improved in terms of noise reduction. Finally, this study enables additional proposals of ANN with Deep Learning models for DBT reconstruction and denoising.
Descrição
Palavras-chave
Idioma
Inglês
Citação
Progress in Biomedical Optics and Imaging - Proceedings of SPIE, v. 11595.
