Articles producció científica> Enginyeria Informàtica i Matemàtiques

Fetal brain tissue annotation and segmentation challenge results

  • Datos identificativos

    Identificador: imarina:9321004
    Handle: https://hdl.handle.net/20.500.11797/imarina9321004
    Autores:
    Payette, KLi, HBde Dumest, PLicandro, RJi, HSiddiquee, MMRXu, DGMyronenko, ALiu, HPei, YCWang, LSPeng, YXie, JYZhang, HQDong, GMFu, HWang, GTRieu, ZKim, DKim, HGKarimi, DGholipour, ATorres, HROliveira, BVilaca, JLLin, YAvisdris, NBen-Zvi, WBen Bashat, DFidon, LAertsen, MVercauteren, TSobotka, DLangs, GAlenya, MVillanueva, MICamara, OFadida, BSJoskowicz, LWeibin, LYi, LXuesong, LMazher, MQayyum, APuig, DKebiri, HZhang, ZLXu, XYWu, DLiao, KLWu, YXChen, JTXu, YZZhao, LVasung, LMenze, BCuadra, MBJakab, A
    Resumen:
    In-utero fetal MRI is emerging as an important tool in the diagnosis and analysis of the developing human brain. Automatic segmentation of the developing fetal brain is a vital step in the quantitative analysis of prenatal neurodevelopment both in the research and clinical context. However, manual segmentation of cerebral structures is time-consuming and prone to error and inter-observer variability. Therefore, we organized the Fetal Tissue Annotation (FeTA) Challenge in 2021 in order to encourage the development of automatic segmentation algorithms on an international level. The challenge utilized FeTA Dataset, an open dataset of fetal brain MRI reconstructions segmented into seven different tissues (external cerebrospinal fluid, gray matter, white matter, ventricles, cerebellum, brainstem, deep gray matter). 20 international teams participated in this challenge, submitting a total of 21 algorithms for evaluation. In this paper, we provide a detailed analysis of the results from both a technical and clinical perspective. All participants relied on deep learning methods, mainly U-Nets, with some variability present in the network architecture, optimization, and image pre- and post-processing. The majority of teams used existing medical imaging deep learning frameworks. The main differences between the submissions were the fine tuning done during training, and the specific pre- and post-processing steps performed. The challenge results showed that almost all submissions performed similarly. Four of the top five teams used ensemble learning methods. However, one team's algorithm performed significantly superior to the other submissions, and consisted of an asymmetrical U-Net network architecture. This paper provides a first of its kind benchmark for future automatic multi

  • Otros:

    Autor según el artículo: Payette, K; Li, HB; de Dumest, P; Licandro, R; Ji, H; Siddiquee, MMR; Xu, DG; Myronenko, A; Liu, H; Pei, YC; Wang, LS; Peng, Y; Xie, JY; Zhang, HQ; Dong, GM; Fu, H; Wang, GT; Rieu, Z; Kim, D; Kim, HG; Karimi, D; Gholipour, A; Torres, HR; Oliveira, B; Vilaca, JL; Lin, Y; Avisdris, N; Ben-Zvi, W; Ben Bashat, D; Fidon, L; Aertsen, M; Vercauteren, T; Sobotka, D; Langs, G; Alenya, M; Villanueva, MI; Camara, O; Fadida, BS; Joskowicz, L; Weibin, L; Yi, L; Xuesong, L; Mazher, M; Qayyum, A; Puig, D; Kebiri, H; Zhang, ZL; Xu, XY; Wu, D; Liao, KL; Wu, YX; Chen, JT; Xu, YZ; Zhao, L; Vasung, L; Menze, B; Cuadra, MB; Jakab, A
    Departamento: Enginyeria Informàtica i Matemàtiques
    Autor/es de la URV: Mazher, Moona / Puig Valls, Domènec Savi
    Palabras clave: Super-resolution reconstructions Multi-class image segmentation Mri Fetal brain mri Congenital disorders super-resolution reconstructions myelomeningocele fetuses fetal brain mri congenital disorders atlas
    Resumen: In-utero fetal MRI is emerging as an important tool in the diagnosis and analysis of the developing human brain. Automatic segmentation of the developing fetal brain is a vital step in the quantitative analysis of prenatal neurodevelopment both in the research and clinical context. However, manual segmentation of cerebral structures is time-consuming and prone to error and inter-observer variability. Therefore, we organized the Fetal Tissue Annotation (FeTA) Challenge in 2021 in order to encourage the development of automatic segmentation algorithms on an international level. The challenge utilized FeTA Dataset, an open dataset of fetal brain MRI reconstructions segmented into seven different tissues (external cerebrospinal fluid, gray matter, white matter, ventricles, cerebellum, brainstem, deep gray matter). 20 international teams participated in this challenge, submitting a total of 21 algorithms for evaluation. In this paper, we provide a detailed analysis of the results from both a technical and clinical perspective. All participants relied on deep learning methods, mainly U-Nets, with some variability present in the network architecture, optimization, and image pre- and post-processing. The majority of teams used existing medical imaging deep learning frameworks. The main differences between the submissions were the fine tuning done during training, and the specific pre- and post-processing steps performed. The challenge results showed that almost all submissions performed similarly. Four of the top five teams used ensemble learning methods. However, one team's algorithm performed significantly superior to the other submissions, and consisted of an asymmetrical U-Net network architecture. This paper provides a first of its kind benchmark for future automatic multi-tissue segmentation algorithms for the developing human brain in utero.
    Áreas temáticas: Radiology, nuclear medicine and imaging Radiology, nuclear medicine & medical imaging Radiological and ultrasound technology Materiais Health informatics Engineering, biomedical Engenharias iv Computer vision and pattern recognition Computer science, interdisciplinary applications Computer science, artificial intelligence Computer graphics and computer-aided design Ciência da computação
    Acceso a la licencia de uso: https://creativecommons.org/licenses/by/3.0/es/
    Direcció de correo del autor: moona.mazher@estudiants.urv.cat domenec.puig@urv.cat
    Identificador del autor: 0000-0003-4444-5776 0000-0002-0562-4205
    Fecha de alta del registro: 2024-08-03
    Versión del articulo depositado: info:eu-repo/semantics/publishedVersion
    Enlace a la fuente original: https://www.sciencedirect.com/science/article/pii/S1361841523000932
    URL Documento de licencia: https://repositori.urv.cat/ca/proteccio-de-dades/
    Referencia al articulo segun fuente origial: Medical Image Analysis. 88
    Referencia de l'ítem segons les normes APA: Payette, K; Li, HB; de Dumest, P; Licandro, R; Ji, H; Siddiquee, MMR; Xu, DG; Myronenko, A; Liu, H; Pei, YC; Wang, LS; Peng, Y; Xie, JY; Zhang, HQ; Do (2023). Fetal brain tissue annotation and segmentation challenge results. Medical Image Analysis, 88(), -. DOI: 10.1016/j.media.2023.102833
    DOI del artículo: 10.1016/j.media.2023.102833
    Entidad: Universitat Rovira i Virgili
    Año de publicación de la revista: 2023
    Tipo de publicación: Journal Publications

  • Palabras clave:

    Computer Graphics and Computer-Aided Design,Computer Science, Artificial Intelligence,Computer Science, Interdisciplinary Applications,Computer Vision and Pattern Recognition,Engineering, Biomedical,Health Informatics,Radiological and Ultrasound Technology,Radiology, Nuclear Medicine & Medical Imaging,Radiology, Nuclear Medicine and Imagin
    Super-resolution reconstructions
    Multi-class image segmentation
    Mri
    Fetal brain mri
    Congenital disorders
    super-resolution reconstructions
    myelomeningocele
    fetuses
    fetal brain mri
    congenital disorders
    atlas
    Radiology, nuclear medicine and imaging
    Radiology, nuclear medicine & medical imaging
    Radiological and ultrasound technology
    Materiais
    Health informatics
    Engineering, biomedical
    Engenharias iv
    Computer vision and pattern recognition
    Computer science, interdisciplinary applications
    Computer science, artificial intelligence
    Computer graphics and computer-aided design
    Ciência da computação

  • Documentos:

    DocumentPrincipal

  • Cerca a google

    Search to google scholar