White Matter, Gray Matter and Cerebrospinal Fluid Segmentation from Brain 3D MRI Using B-UNET

• Autor(a) principal: Tran Anh Tuan
• Data de Publicação: 2019
• Outros Autores: Pham The Bao, Jin Young Kim, João Manuel R. S. Tavares
• Tipo de documento: Livro
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/124729
• topic: Ciências Tecnológicas, Ciências médicas e da saúde; Technological sciences, Medical and Health sciences

The accurate segmentation of brain tissues in Magnetic Resonance (MR) images is an important step for detection and treatment planning of brain diseases. Among other brain tissues, Gray Matter, White Matter and Cerebrospinal Fluid are commonly segmented for Alzheimer diagnosis purpose. Therefore, different algorithms for segmenting these tissues in MR image scans have been proposed over the years. Nowadays, with the trend of deep learning, many methods are trained to learn important features and extract information from the data leading to very promising segmentation results. In this work, we propose an effective approach to segment three tissues in 3D Brain MR images based on B-UNET. The method is implemented by using the Bitplane method in each convolution of the UNET model. We evaluated the proposed method using two public databases with very promising results. (c) Springer Nature Switzerland AG 2019.