View in EDS PDF Full Text

High grade glioma segmentation in magnetic resonance imaging.

Saved in:
Bibliographic Details
Title: High grade glioma segmentation in magnetic resonance imaging.
Alternate Title: Segmentatión de glioma de alto grado en imágenes de resonancia magnética.
Authors: Vera, Miguel1 m.avera@unisimonbolivar.edu.co, Huérfano, Yoleidy2, Martinez, Luis Javier3, Contreras, Yudith1, Salazar, Williams4, Vera, Maria Isabel4, Valbuena, Oscar5, Borrero, Maryury1, Hernéndez, Carlos1, Barrera, Doris1, Molina, Angel Valentin3, Salazar, Juan1, Gelvez, Elkin1, Séenz, Frank6, Arias, Yeny1
Source: Revista Latinoamericana de Hipertensión. 2018, Vol. 13 Issue 4, p323-329. 7p.
Subjects: MAGNETIC resonance imaging, BRAIN imaging, IMAGE processing, BRAIN tumors, GLIOMAS
Abstract (English): Through this work we propose a computational technique for the segmentation of magnetic resonance images (MRI) of a brain tumor, identified as high grade glioma (HGG), specifically grade III anaplastic astrocytoma. This technique consists of 3 stages developed in the threedimensional domain. They are: pre-processing, segmentation and post-processing. The pre-processing stage uses a thresholding technique, morphological erosion filter (MEF), in gray scale, followed by a median filter and a gradient magnitude algorithm. On the other hand, in order to obtain a HGG preliminary segmentation, during the segmentation stage a clustering algorithm called region growing (RG) is implemented and it is applied to the preprocessed images. The RG requires, for its initialization, a seed voxel whose coordinates are obtained, automatically, through the training and validation of an intelligent operator based on support vector machines (SVM). Due to the high sensitivity of the RG to the location of the seed, the SVM is implemented as a highly selective binary classifier. During the post-processing stage, a morphological dilation filter is applied to preliminary segmentation generated by RG. The percent relative error (PrE) is considered by comparing the segmentations of the HGG, generated manually by a neuro-oncologist, with the dilated segmentations of the HGG, obtained automatically. The combination of parameters linked to the lowest PrE, allows establishing the optimal parameters of each computational algorithms that make up the proposed computational technique. The obtained results allow reporting a PrE of 11.10%, which indicates a good correlation between the manual segmentations and those produced by the computational technique developed. [ABSTRACT FROM AUTHOR]
Abstract (Spanish): A través de este trabajo se propone una técnica computacional para la segmentacion de un tumor cerebral, identificado como un glioma de alto grado (HGG) de tipo astrocitoma anaplasico de grado III, que esta presente en las imagenes de resonancia magnética (MRI). Esta técnica consta de 3 etapas desarrolladas en el dominio tridimensional. Ellas son: preprocesamiento, segmentacion y postprocesamiento. La etapa de preprocesamiento utiliza una técnica de umbralizacion, un filtro de erosion morfologica (MEF), en escala de grises, seguido de un filtro de mediana y de un algoritmo de magnitud de gradiente. Por otro lado, con el proposito de generar una segmentacion preliminar del HGG, durante la etapa de segmentacion se implementa un algoritmo de agrupamiento, llamado crecimiento de regiones (RG), que se aplica a las imagenes preprocesadas. El RG requiere para su inicializacion la ubicacion de un voxel semilla cuyas coordenadas se obtienen, automaticamente, a través del entrenamiento y la validacion de un operador inteligente basado en maquinas de vectores de soporte (SVM). Debido a la alta sensibilidad del RG a la ubicacion de la semilla, la SVM se implementa como un clasificador binario altamente selectivo. Durante la etapa de post-procesamiento, se aplica un filtro de dilatacion morfologica a la segmentacion preliminar, generada por RG. El error relativo porcentual (PrE) se considera para comparar las segmentaciones de la HGG generadas de forma manual por un neurooncologo, con las segmentaciones dilatadas de la HGG, obtenidas automaticamente. La combinacion de parametros vinculados al PrE mas bajo permite establecer los parametros optimos de cada uno de los algoritmos computacionales que componen la técnica computacional propuesta. Los resultados obtenidos permiten reportar un PrE de 11.10%, lo cual indica una buena correlacion entre las segmentaciones manuales y las producidas por la técnica computacional desarrollada. [ABSTRACT FROM AUTHOR]
Copyright of Revista Latinoamericana de Hipertensión is the property of Revista Latinoamericana de Hipertension and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: MedicLatina
Description
Abstract:Through this work we propose a computational technique for the segmentation of magnetic resonance images (MRI) of a brain tumor, identified as high grade glioma (HGG), specifically grade III anaplastic astrocytoma. This technique consists of 3 stages developed in the threedimensional domain. They are: pre-processing, segmentation and post-processing. The pre-processing stage uses a thresholding technique, morphological erosion filter (MEF), in gray scale, followed by a median filter and a gradient magnitude algorithm. On the other hand, in order to obtain a HGG preliminary segmentation, during the segmentation stage a clustering algorithm called region growing (RG) is implemented and it is applied to the preprocessed images. The RG requires, for its initialization, a seed voxel whose coordinates are obtained, automatically, through the training and validation of an intelligent operator based on support vector machines (SVM). Due to the high sensitivity of the RG to the location of the seed, the SVM is implemented as a highly selective binary classifier. During the post-processing stage, a morphological dilation filter is applied to preliminary segmentation generated by RG. The percent relative error (PrE) is considered by comparing the segmentations of the HGG, generated manually by a neuro-oncologist, with the dilated segmentations of the HGG, obtained automatically. The combination of parameters linked to the lowest PrE, allows establishing the optimal parameters of each computational algorithms that make up the proposed computational technique. The obtained results allow reporting a PrE of 11.10%, which indicates a good correlation between the manual segmentations and those produced by the computational technique developed. [ABSTRACT FROM AUTHOR]
ISSN:18564550