المؤلفون: Senra Filho, Antonio Carlos da Silva

المصدر: Research on Biomedical Engineering. September 2016 32(3)

مصطلحات موضوعية: Relaxometry, Magnetic resonance imaging, Brain phantom, Simulation

الوصف: Introduction Relaxometry images are an important magnetic resonance imaging (MRI) technique in the clinical routine. Many diagnoses are based on the relaxometry maps to infer abnormal state in the tissue characteristic relaxation constant. In order to study the performance of these image processing approaches, a controlled simulated environment is necessary. However, a simulated relaxometry image tool is still lacking. This study proposes a computational anatomical brain phantom for MRI relaxometry images, which aims to offer an easy and flexible toolkit to test different image processing techniques, applied to MRI relaxometry maps in a controlled simulated environment. Methods A pipeline of image processing techniques such as brain extraction, image segmentation, normalization to a common space and signal relaxation decay simulation, were applied to a brain structural ICBM brain template, on both T1 and T2 weighted images, in order to simulate a volumetric brain relaxometry phantom. The FMRIB Software Library (FSL) toolkits were used here as the base image processing needed to all the relaxometry reconstruction. Results All the image processing procedures are performed using automatic algorithms. In addition, different artefact levels can be set from different sources such as Rician noise and radio-frequency inhomogeneity noises. Conclusion The main goal of this project is to help researchers in their future image processing analysis involving MRI relaxometry images, offering reliable and robust brain relaxometry simulation modelling. Furthermore, the entire pipeline is open-source, which provides a wide collaboration between researchers who may want to improve the software and its functionality.

وصف الملف: text/html

الوصول الحر: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2446-47402016000300301Test

المؤلفون: Antonio Carlos da Silva Senra Filho

المصدر: Research on Biomedical Engineering v.32 n.3 2016
Research on Biomedical Engineering
Sociedade Brasileira de Engenharia Biomédica (SBEB)
instacron:SBEB
Research on Biomedical Engineering, Vol 32, Iss 3, Pp 301-305
Research on Biomedical Engineering, Volume: 32, Issue: 3, Pages: 301-305, Published: 10 OCT 2016

مصطلحات موضوعية: Relaxometry, Computer science, lcsh:Biotechnology, Biomedical Engineering, Normalization (image processing), ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Image processing, Imaging phantom, 030218 nuclear medicine & medical imaging, 03 medical and health sciences, 0302 clinical medicine, Software, Magnetic resonance imaging, lcsh:TP248.13-248.65, Computer vision, lcsh:R5-920, business.industry, Brain phantom, Image segmentation, Pipeline (software), FMRIB Software Library, Artificial intelligence, lcsh:Medicine (General), business, 030217 neurology & neurosurgery, Simulation

الوصف: Introduction Relaxometry images are an important magnetic resonance imaging (MRI) technique in the clinical routine. Many diagnoses are based on the relaxometry maps to infer abnormal state in the tissue characteristic relaxation constant. In order to study the performance of these image processing approaches, a controlled simulated environment is necessary. However, a simulated relaxometry image tool is still lacking. This study proposes a computational anatomical brain phantom for MRI relaxometry images, which aims to offer an easy and flexible toolkit to test different image processing techniques, applied to MRI relaxometry maps in a controlled simulated environment. Methods A pipeline of image processing techniques such as brain extraction, image segmentation, normalization to a common space and signal relaxation decay simulation, were applied to a brain structural ICBM brain template, on both T1 and T2 weighted images, in order to simulate a volumetric brain relaxometry phantom. The FMRIB Software Library (FSL) toolkits were used here as the base image processing needed to all the relaxometry reconstruction. Results All the image processing procedures are performed using automatic algorithms. In addition, different artefact levels can be set from different sources such as Rician noise and radio-frequency inhomogeneity noises. Conclusion The main goal of this project is to help researchers in their future image processing analysis involving MRI relaxometry images, offering reliable and robust brain relaxometry simulation modelling. Furthermore, the entire pipeline is open-source, which provides a wide collaboration between researchers who may want to improve the software and its functionality.

وصف الملف: text/html

الوصول الحر: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e11f704a9eb51743c752b6674f316258Test
http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2446-47402016000300301Test