Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net

MRI images are visually inspected by domain experts for the analysis and quantification of the tumorous tissues. Due to the large volumetric data, manual reporting on the images is subjective, cumbersome, and error prone. To address these problems, automatic image analysis tools are employed for tum...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Faizad Ullah, Shahab U. Ansari, Muhammad Hanif, Mohamed Arselene Ayari, Muhammad Enamul Hoque Chowdhury, Amith Abdullah Khandakar, Muhammad Salman Khan
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/a3a21b2648b641f7bead0236f81207a2
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:a3a21b2648b641f7bead0236f81207a2
record_format dspace
spelling oai:doaj.org-article:a3a21b2648b641f7bead0236f81207a22021-11-25T18:57:10ZBrain MR Image Enhancement for Tumor Segmentation Using 3D U-Net10.3390/s212275281424-8220https://doaj.org/article/a3a21b2648b641f7bead0236f81207a22021-11-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/22/7528https://doaj.org/toc/1424-8220MRI images are visually inspected by domain experts for the analysis and quantification of the tumorous tissues. Due to the large volumetric data, manual reporting on the images is subjective, cumbersome, and error prone. To address these problems, automatic image analysis tools are employed for tumor segmentation and other subsequent statistical analysis. However, prior to the tumor analysis and quantification, an important challenge lies in the pre-processing. In the present study, permutations of different pre-processing methods are comprehensively investigated. In particular, the study focused on Gibbs ringing artifact removal, bias field correction, intensity normalization, and adaptive histogram equalization (AHE). The pre-processed MRI data is then passed onto 3D U-Net for automatic segmentation of brain tumors. The segmentation results demonstrated the best performance with the combination of two techniques, i.e., Gibbs ringing artifact removal and bias-field correction. The proposed technique achieved mean dice score metrics of 0.91, 0.86, and 0.70 for the whole tumor, tumor core, and enhancing tumor, respectively. The testing mean dice scores achieved by the system are 0.90, 0.83, and 0.71 for the whole tumor, core tumor, and enhancing tumor, respectively. The novelty of this work concerns a robust pre-processing sequence for improving the segmentation accuracy of MR images. The proposed method overcame the testing dice scores of the state-of-the-art methods. The results are benchmarked with the existing techniques used in the Brain Tumor Segmentation Challenge (BraTS) 2018 challenge.Faizad UllahShahab U. AnsariMuhammad HanifMohamed Arselene AyariMuhammad Enamul Hoque ChowdhuryAmith Abdullah KhandakarMuhammad Salman KhanMDPI AGarticlebrain tumor segmentationdeep learningGibbs ringing artifactimage enhancementmedical image processingChemical technologyTP1-1185ENSensors, Vol 21, Iss 7528, p 7528 (2021)
institution DOAJ
collection DOAJ
language EN
topic brain tumor segmentation
deep learning
Gibbs ringing artifact
image enhancement
medical image processing
Chemical technology
TP1-1185
spellingShingle brain tumor segmentation
deep learning
Gibbs ringing artifact
image enhancement
medical image processing
Chemical technology
TP1-1185
Faizad Ullah
Shahab U. Ansari
Muhammad Hanif
Mohamed Arselene Ayari
Muhammad Enamul Hoque Chowdhury
Amith Abdullah Khandakar
Muhammad Salman Khan
Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
description MRI images are visually inspected by domain experts for the analysis and quantification of the tumorous tissues. Due to the large volumetric data, manual reporting on the images is subjective, cumbersome, and error prone. To address these problems, automatic image analysis tools are employed for tumor segmentation and other subsequent statistical analysis. However, prior to the tumor analysis and quantification, an important challenge lies in the pre-processing. In the present study, permutations of different pre-processing methods are comprehensively investigated. In particular, the study focused on Gibbs ringing artifact removal, bias field correction, intensity normalization, and adaptive histogram equalization (AHE). The pre-processed MRI data is then passed onto 3D U-Net for automatic segmentation of brain tumors. The segmentation results demonstrated the best performance with the combination of two techniques, i.e., Gibbs ringing artifact removal and bias-field correction. The proposed technique achieved mean dice score metrics of 0.91, 0.86, and 0.70 for the whole tumor, tumor core, and enhancing tumor, respectively. The testing mean dice scores achieved by the system are 0.90, 0.83, and 0.71 for the whole tumor, core tumor, and enhancing tumor, respectively. The novelty of this work concerns a robust pre-processing sequence for improving the segmentation accuracy of MR images. The proposed method overcame the testing dice scores of the state-of-the-art methods. The results are benchmarked with the existing techniques used in the Brain Tumor Segmentation Challenge (BraTS) 2018 challenge.
format article
author Faizad Ullah
Shahab U. Ansari
Muhammad Hanif
Mohamed Arselene Ayari
Muhammad Enamul Hoque Chowdhury
Amith Abdullah Khandakar
Muhammad Salman Khan
author_facet Faizad Ullah
Shahab U. Ansari
Muhammad Hanif
Mohamed Arselene Ayari
Muhammad Enamul Hoque Chowdhury
Amith Abdullah Khandakar
Muhammad Salman Khan
author_sort Faizad Ullah
title Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
title_short Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
title_full Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
title_fullStr Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
title_full_unstemmed Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net
title_sort brain mr image enhancement for tumor segmentation using 3d u-net
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/a3a21b2648b641f7bead0236f81207a2
work_keys_str_mv AT faizadullah brainmrimageenhancementfortumorsegmentationusing3dunet
AT shahabuansari brainmrimageenhancementfortumorsegmentationusing3dunet
AT muhammadhanif brainmrimageenhancementfortumorsegmentationusing3dunet
AT mohamedarseleneayari brainmrimageenhancementfortumorsegmentationusing3dunet
AT muhammadenamulhoquechowdhury brainmrimageenhancementfortumorsegmentationusing3dunet
AT amithabdullahkhandakar brainmrimageenhancementfortumorsegmentationusing3dunet
AT muhammadsalmankhan brainmrimageenhancementfortumorsegmentationusing3dunet
_version_ 1718410544656941056