Super-resolution in brain Diffusion Weighted Imaging (DWI)

Abstract. Diffusion Weighted (DW) imaging has proven to be useful at analysing brain architecture as well as at establishing brain tract organization and neuronal connectivity. However, an actual clinical use of DW images is currently limited by a series of acquisition artifacts, among them the partial volume effect (PVE) that may completely alter the spatial resolution and therefore the visualization of microanatomical details. In this work, a new superresolution method will be presented, taking advantage of the redundant structural patterns that shape the brain. The proposed method couples low-high resolution information and explores different directional spaces that might exploit the spectral content of the DW images. A comparison of this proposal with a classical image interpolation method demostrates an improvement of about 3 dB when using the typical PSNR.

• Autores: Tarquino González, Jonathan Steve
• Fecha de publicación: 2014
• Institución: Universidad Nacional de Colombia
• Repositorio: Universidad Nacional de Colombia
• Idioma: spa
• OAI Identifier: oai:repositorio.unal.edu.co:unal/54123
• Acceso en línea: https://repositorio.unal.edu.co/handle/unal/54123; http://bdigital.unal.edu.co/48958/
• Palabra clave: 61 Ciencias médicas; Medicina / Medicine and health; 62 Ingeniería y operaciones afines / Engineering; Super-resolution; Image processing; Diffusion Weighted Imaging; Shearlet transform; Multiscale representation; Sparse representation; Super-resoluci´on,; Imágenes de Diffusión Ponderada (DWI); Transformada Shearlet; Representaciones multi-escala; Representaciones sparse
• Rights: openAccess
• License: Atribución-NoComercial 4.0 Internacional