Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images

A brain tumor is a mass of abnormal cells that grow uncontrollably in the brain or central spine. Glioblastoma is the most common malignant primary brain tumor, accounting for over 48.6 \% of all malignant primary brain tumors, and 14.5 \% of all primary brain tumors. Regardless of the diagnosis, th...

ver descrição completa

Na minha lista:
Detalhes bibliográficos
Autor principal: Tene Hurtado, Daniela Verónica (author)
Formato: bachelorThesis
Idioma:eng
Publicado em: 2021
Assuntos:
Acesso em linha:http://repositorio.yachaytech.edu.ec/handle/123456789/460
Tags: Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
_version_ 1840070391936057344
author Tene Hurtado, Daniela Verónica
author_facet Tene Hurtado, Daniela Verónica
author_role author
collection Repositorio Universidad Yachay Tech
dc.contributor.none.fl_str_mv Almeida Galárraga, Diego Alfonso
dc.creator.none.fl_str_mv Tene Hurtado, Daniela Verónica
dc.date.none.fl_str_mv 2021-12
2022-01-10T10:23:46Z
2022-01-10T10:23:46Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://repositorio.yachaytech.edu.ec/handle/123456789/460
dc.language.none.fl_str_mv eng
dc.publisher.none.fl_str_mv Universidad de Investigación de Tecnología Experimental Yachay
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
dc.source.none.fl_str_mv reponame:Repositorio Universidad Yachay Tech
instname:Universidad Yachay Tech
instacron:Yachay
dc.subject.none.fl_str_mv Glioma
Segmentación de tumores cerebrales
U-Net
Redes neuronales
Brain tumor segmentation
Multimodal MRI
BraTS dataset
Neural networks
dc.title.none.fl_str_mv Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
dc.type.none.fl_str_mv info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/bachelorThesis
description A brain tumor is a mass of abnormal cells that grow uncontrollably in the brain or central spine. Glioblastoma is the most common malignant primary brain tumor, accounting for over 48.6 \% of all malignant primary brain tumors, and 14.5 \% of all primary brain tumors. Regardless of the diagnosis, the morbidity and mortality represented by a tumor located in the brain remain significant. They have a 5-year relative survival rate of about 7.2 \%, which means that only 7.2 \% of people diagnosed with glioblastoma will live 5 years after diagnosis; the median survival time is about 8 months. Detecting the presence of brain tumors using magnetic resonance images in a rapid, precise, and reproducible manner is a difficult but still necessary task. Brain tumor segmentation is a key factor for the analysis of gliomas, it involves the delineation of the different tumor regions including peritumoral edema, enhancing tumor, and tumor core. Currently, different methodologies in the literature implement the tumor segmentation technique to improve the diagnosis and treatment plan. The proposed methodology consists of an automated segmentation of brain tumors with the use of neural networks in particular U-Net and Attention U-Net, which are mainly used for fast and accurate segmentation of biomedical images. The BraTS 2020 dataset is used in this study to test the egmentation performance of the suggested technique. An accuracy of 0.9950 was obtained for both models, and a sensitivity of 0.9931 and 0.9891 for the U-Net and Attention U-Net models. A peritumoral edema, enhancing tumor, and tumor core dice similarity coefficient of 0.8453, 0.6950, and 0.7429 respectively has been achieved, for the U-Net model. For the Attention U-Net model, a dice score of 0.8829 0.7233, and 0.8090 were obtained. The study presents and discusses further quantitative and qualitative assessments. Results show that both approaches have considerable potential and can be employed in clinical practice in the segmentation of various sub-regions of brain tumors.
eu_rights_str_mv openAccess
format bachelorThesis
id Yachay_eb67aeb02c68179d18f51bafb88ef1dd
instacron_str Yachay
institution Yachay
instname_str Universidad Yachay Tech
language eng
network_acronym_str Yachay
network_name_str Repositorio Universidad Yachay Tech
oai_identifier_str oai:repositorio.yachaytech.edu.ec:123456789/460
publishDate 2021
publisher.none.fl_str_mv Universidad de Investigación de Tecnología Experimental Yachay
reponame_str Repositorio Universidad Yachay Tech
repository.mail.fl_str_mv .
repository.name.fl_str_mv Repositorio Universidad Yachay Tech - Universidad Yachay Tech
repository_id_str 10284
spelling Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain imagesTene Hurtado, Daniela VerónicaGliomaSegmentación de tumores cerebralesU-NetRedes neuronalesBrain tumor segmentationMultimodal MRIBraTS datasetNeural networksA brain tumor is a mass of abnormal cells that grow uncontrollably in the brain or central spine. Glioblastoma is the most common malignant primary brain tumor, accounting for over 48.6 \% of all malignant primary brain tumors, and 14.5 \% of all primary brain tumors. Regardless of the diagnosis, the morbidity and mortality represented by a tumor located in the brain remain significant. They have a 5-year relative survival rate of about 7.2 \%, which means that only 7.2 \% of people diagnosed with glioblastoma will live 5 years after diagnosis; the median survival time is about 8 months. Detecting the presence of brain tumors using magnetic resonance images in a rapid, precise, and reproducible manner is a difficult but still necessary task. Brain tumor segmentation is a key factor for the analysis of gliomas, it involves the delineation of the different tumor regions including peritumoral edema, enhancing tumor, and tumor core. Currently, different methodologies in the literature implement the tumor segmentation technique to improve the diagnosis and treatment plan. The proposed methodology consists of an automated segmentation of brain tumors with the use of neural networks in particular U-Net and Attention U-Net, which are mainly used for fast and accurate segmentation of biomedical images. The BraTS 2020 dataset is used in this study to test the egmentation performance of the suggested technique. An accuracy of 0.9950 was obtained for both models, and a sensitivity of 0.9931 and 0.9891 for the U-Net and Attention U-Net models. A peritumoral edema, enhancing tumor, and tumor core dice similarity coefficient of 0.8453, 0.6950, and 0.7429 respectively has been achieved, for the U-Net model. For the Attention U-Net model, a dice score of 0.8829 0.7233, and 0.8090 were obtained. The study presents and discusses further quantitative and qualitative assessments. Results show that both approaches have considerable potential and can be employed in clinical practice in the segmentation of various sub-regions of brain tumors.Un tumor cerebral es una masa de células anormales que crece de forma incontrolada en el cerebro o en la columna vertebral. El glioblastoma es el tumor cerebral primario maligno más frecuente, ya que representa más del 48.6 \% de todos los tumores cerebrales primarios malignos, y el 14.5 \% de todos los tumores cerebrales primrios. Independientemente del diagnóstico, la morbilidad y la mortalidad que representa un tumor localizado en el cerebro siguen siendo importantes. La tasa de supervivencia relativa a los 5 años es de aproximadamente el 7,2 \%, lo que significa que sólo el 7,2 \% de las personas diagnosticadas con glioblastoma vivirán 5 años después del diagnóstico; la supervivencia media es de aproximadamente 8 meses. Detectar la presencia de tumores cerebrales mediante imágenes de resonancia magnética de forma rápida, precisa y reproducible es una tarea difícil pero necesaria. La segmentación de los tumores cerebrales es un factor clave para el análisis de los gliomas, implica la delineación de las diferentes regiones tumorales, incluyendo el edema peritumoral, el tumor realzante y el núcleo del tumor. Actualmente, existen diferentes metodologías en la literature que implementan la técnica de segmentación tumoral para mejorar el diagnóstico y el plan de tratamiento. La metodología propuesta consiste en una egmentación automatizada de tumores cerebrales con el uso de redes neuronales en particular U-Net y Attention U-Net, ambas arquitecturas son altamente empleadas para la segmentación rápida y precisa de imágenes biomédicas. Estos metodos se evaluarón en elconjunto de datos BraTS 2020. Se obtuvó una precisión de 0,9950 para ambos modelos, y una sensibilidad de 0,9931 y 0,9891 para los modelos U-Net y Attention U-Net, respectivamente. Asimismo se valoró el coeficiente de Sorensen, el modelo U-Net obtuvó valores para el edema peritumoral, tumor en aumento y núcleo del tumor de 0,8453, 0,695 y 0,7429 respectivamente. De igual manera, para el modelo de Attention U-Net, se alcanzó una un coeficiente de 0,8829, 0,7233 y 0,8090. El estudio presenta y analiza otras evaluaciones cuantitativas y cualitativas. Los resultados muestran que ambos enfoques tienen un potencial considerable y pueden emplearse en la práctica clínica en la segmentación de varias subregiones de tumores cerebrales.Ingeniero/a Biomédico/aUniversidad de Investigación de Tecnología Experimental YachayAlmeida Galárraga, Diego Alfonso2022-01-10T10:23:46Z2022-01-10T10:23:46Z2021-12info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/bachelorThesisapplication/pdfhttp://repositorio.yachaytech.edu.ec/handle/123456789/460enginfo:eu-repo/semantics/openAccessreponame:Repositorio Universidad Yachay Techinstname:Universidad Yachay Techinstacron:Yachay2025-07-08T17:55:27Zoai:repositorio.yachaytech.edu.ec:123456789/460Institucionalhttps://repositorio.yachaytech.edu.ec/Universidad públicahttps://www.yachaytech.edu.ec/https://repositorio.yachaytech.edu.ec/oaiEcuador...opendoar:102842025-07-08T17:55:27falseInstitucionalhttps://repositorio.yachaytech.edu.ec/Universidad públicahttps://www.yachaytech.edu.ec/https://repositorio.yachaytech.edu.ec/oai.Ecuador...opendoar:102842025-07-08T17:55:27Repositorio Universidad Yachay Tech - Universidad Yachay Techfalse
spellingShingle Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
Tene Hurtado, Daniela Verónica
Glioma
Segmentación de tumores cerebrales
U-Net
Redes neuronales
Brain tumor segmentation
Multimodal MRI
BraTS dataset
Neural networks
status_str publishedVersion
title Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
title_full Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
title_fullStr Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
title_full_unstemmed Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
title_short Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
title_sort Brain tumor segmentation based on 2D U-Net using MRI multi-modalities brain images
topic Glioma
Segmentación de tumores cerebrales
U-Net
Redes neuronales
Brain tumor segmentation
Multimodal MRI
BraTS dataset
Neural networks
url http://repositorio.yachaytech.edu.ec/handle/123456789/460