Automatic segmentation of a cerebral glioblastoma using a smart computational technique

We propose an intelligent computational technique for the image segmentation of a type IV brain tumor, identified as multiform glioblastoma (MGB), which is present in multi-layer computed tomography images. This technique consists of 3 stages developed in the three-dimensional domain. They are: pre-processing, segmentation and validation. During the validation stage, the Dice coefficient (Dc) is considered in order to compare the segmentations of the MGB, obtained automatically, with the segmentations of the MGB generated manually, by a neuro-oncologist. The combination of parameters linked to the highest Dc, allows to establish the optimal parameters of each of the computational algorithms that make up the proposed nonlinear technique. The obtained results allow to report a Dc higher than 0.88, validating a good correlation between the manual segmentations and those produced by the computational technique developed.

• Autores: Vera, Miguel; Huérfano, Yoleidy; Valbuena, Oscar; Hoyos, Diego; Arias, Yeni; Contreras, Yudith; Salazar, Williams; Vera, María Isabel; Borrero, Maryury; Vivas, Marisela; Hernández, Carlos; Barrera, Doris; Molina, Ángel Valentín; Martínez, Luis Javier; Salazar, Juan; Gelvez, Elkin
• Fecha de publicación: 2018
• Institución: Universidad Simón Bolívar
• Repositorio: Repositorio Digital USB
• Idioma: eng
• OAI Identifier: oai:bonga.unisimon.edu.co:20.500.12442/2524
• Acceso en línea: http://hdl.handle.net/20.500.12442/2524
• Palabra clave: Brain Tomography; Cerebral Tumor; Glioblastoma; Intelligent Computational Technique; Segmentation; Tomografía cerebral; Tumor cerebral; Glioblastoma; Técnica computacional inteligente; Segmentación
• License: http://purl.org/coar/access_right/c_abf2