TY - GEN
T1 - Gaussian processes for slice-based super-resolution MR images
AU - Cardona, Hernán Darío Vargas
AU - López-Lopera, Andrés F.
AU - Orozco, Álvaro A.
AU - Álvarez, Mauricio A.
AU - Tamames, Juan Antonio Hernández
AU - Malpica, Norberto
N1 - Publisher Copyright:
© Springer International Publishing Switzerland 2015.
PY - 2015
Y1 - 2015
N2 - Magnetic resonance imaging (MRI) is a medical technique used in radiology to obtain anatomical images of healthy and pathological tissues. Due to hardware limitations and clinical protocols, MRI data are often acquired with low-resolution. For this reason, the scientific community has been developing super-resolution (SR) methodologies in order to enhance spatial resolution through post-processing of 2D multi-slice images. The enhancement of spatial resolution in magnetic resonance (MR) images improves clinical procedures such as tissue segmentation, registration and disease diagnosis. Several methods to perform SR-MR images have been proposed. However, they present different drawbacks: sensitivity to noise, high computational cost, and complex optimization algorithms. In this paper, we develop a supervised learning methodology to perform SR-MR images using a patch-based Gaussian process regression (GPR) method. We compare our approach with nearest-neighbor interpolation, B-splines and a SR-GPR scheme based on nearest-neighbors. We test our SR-GPR algorithm in MRIT1 and MRI-T2 studies, evaluating the performance through error metrics and morphological validation (tissue segmentation). Results obtained with our methodology outperform the other alternatives for all validation protocols.
AB - Magnetic resonance imaging (MRI) is a medical technique used in radiology to obtain anatomical images of healthy and pathological tissues. Due to hardware limitations and clinical protocols, MRI data are often acquired with low-resolution. For this reason, the scientific community has been developing super-resolution (SR) methodologies in order to enhance spatial resolution through post-processing of 2D multi-slice images. The enhancement of spatial resolution in magnetic resonance (MR) images improves clinical procedures such as tissue segmentation, registration and disease diagnosis. Several methods to perform SR-MR images have been proposed. However, they present different drawbacks: sensitivity to noise, high computational cost, and complex optimization algorithms. In this paper, we develop a supervised learning methodology to perform SR-MR images using a patch-based Gaussian process regression (GPR) method. We compare our approach with nearest-neighbor interpolation, B-splines and a SR-GPR scheme based on nearest-neighbors. We test our SR-GPR algorithm in MRIT1 and MRI-T2 studies, evaluating the performance through error metrics and morphological validation (tissue segmentation). Results obtained with our methodology outperform the other alternatives for all validation protocols.
UR - https://www.scopus.com/pages/publications/84952802545
U2 - 10.1007/978-3-319-27863-6_65
DO - 10.1007/978-3-319-27863-6_65
M3 - Conference contribution
AN - SCOPUS:84952802545
SN - 9783319278629
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 692
EP - 701
BT - Advances in Visual Computing - 11th International Symposium, ISVC 2015, Proceedings
A2 - Parvin, Bahram
A2 - Koracin, Darko
A2 - Feris, Rogerio
A2 - Weber, Gunther
A2 - Pavlidis, Ioannis
A2 - McGraw, Tim
A2 - Kopper, Regis
A2 - Ye, Zhao
A2 - Ragan, Eric
A2 - Bebis, George
A2 - Elendt, Mark
A2 - Boyle, Richard
PB - Springer Verlag
T2 - 11th International Symposium on Advances in Visual Computing , ISVC 2015
Y2 - 14 December 2015 through 16 December 2015
ER -