Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases
Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, and these common procedures are usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-04-01
|
| Series: | NeuroImage |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1053811921000112 |
| _version_ | 1831692472933154816 |
|---|---|
| author | Riccardo De Feo Artem Shatillo Alejandra Sierra Juan Miguel Valverde Olli Gröhn Federico Giove Jussi Tohka |
| author_facet | Riccardo De Feo Artem Shatillo Alejandra Sierra Juan Miguel Valverde Olli Gröhn Federico Giove Jussi Tohka |
| author_sort | Riccardo De Feo |
| collection | DOAJ |
| description | Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, and these common procedures are usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher segmentation accuracy than state-of-the-art multi-atlas segmentation methods with an inference time of 0.35 s and no pre-processing requirements.We trained and validated MU-Net on 128 T2-weighted mouse MRI volumes as well as on the publicly available MRM NeAT dataset of 10 MRI volumes. We tested MU-Net with an unusually large dataset combining several independent studies consisting of 1782 mouse brain MRI volumes of both healthy and Huntington animals, and measured average Dice scores of 0.906 (striati), 0.937 (cortex), and 0.978 (brain mask). Further, we explored the effectiveness of our network in the presence of different architectural features, including skip connections and recently proposed framing connections, and the effects of the age range of the training set animals.These high evaluation scores demonstrate that MU-Net is a powerful tool for segmentation and skull-stripping, decreasing inter and intra-rater variability of manual segmentation. The MU-Net code and the trained model are publicly available at https://github.com/Hierakonpolis/MU-Net. |
| first_indexed | 2024-12-20T11:54:15Z |
| format | Article |
| id | doaj.art-8bd127d8eac04df9b8c4ea27edf1cfa4 |
| institution | Directory Open Access Journal |
| issn | 1095-9572 |
| language | English |
| last_indexed | 2024-12-20T11:54:15Z |
| publishDate | 2021-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage |
| spelling | doaj.art-8bd127d8eac04df9b8c4ea27edf1cfa42022-12-21T19:41:42ZengElsevierNeuroImage1095-95722021-04-01229117734Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databasesRiccardo De Feo0Artem Shatillo1Alejandra Sierra2Juan Miguel Valverde3Olli Gröhn4Federico Giove5Jussi Tohka6Corresponding author at: Sapienza Università di Roma, 00184 Rome, Italy.; Sapienza Università di Roma, Rome 00184, Italy; Centro Fermi–Museo Storico della Fisica e Centro Studi e Ricerche Enrico Fermi, Rome 00184, Italy; A.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio 70210, FinlandCharles River Discovery Services, Kuopio, FinlandA.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio 70210, FinlandA.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio 70210, FinlandA.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio 70210, FinlandCentro Fermi–Museo Storico della Fisica e Centro Studi e Ricerche Enrico Fermi, Rome 00184, Italy; Fondazione Santa Lucia IRCCS, Rome 00179, ItalyA.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio 70210, FinlandSkull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, and these common procedures are usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher segmentation accuracy than state-of-the-art multi-atlas segmentation methods with an inference time of 0.35 s and no pre-processing requirements.We trained and validated MU-Net on 128 T2-weighted mouse MRI volumes as well as on the publicly available MRM NeAT dataset of 10 MRI volumes. We tested MU-Net with an unusually large dataset combining several independent studies consisting of 1782 mouse brain MRI volumes of both healthy and Huntington animals, and measured average Dice scores of 0.906 (striati), 0.937 (cortex), and 0.978 (brain mask). Further, we explored the effectiveness of our network in the presence of different architectural features, including skip connections and recently proposed framing connections, and the effects of the age range of the training set animals.These high evaluation scores demonstrate that MU-Net is a powerful tool for segmentation and skull-stripping, decreasing inter and intra-rater variability of manual segmentation. The MU-Net code and the trained model are publicly available at https://github.com/Hierakonpolis/MU-Net.http://www.sciencedirect.com/science/article/pii/S1053811921000112MRIBrainSegmentationDeep learningU-NetMice |
| spellingShingle | Riccardo De Feo Artem Shatillo Alejandra Sierra Juan Miguel Valverde Olli Gröhn Federico Giove Jussi Tohka Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases NeuroImage MRI Brain Segmentation Deep learning U-Net Mice |
| title | Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases |
| title_full | Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases |
| title_fullStr | Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases |
| title_full_unstemmed | Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases |
| title_short | Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases |
| title_sort | automated joint skull stripping and segmentation with multi task u net in large mouse brain mri databases |
| topic | MRI Brain Segmentation Deep learning U-Net Mice |
| url | http://www.sciencedirect.com/science/article/pii/S1053811921000112 |
| work_keys_str_mv | AT riccardodefeo automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT artemshatillo automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT alejandrasierra automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT juanmiguelvalverde automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT olligrohn automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT federicogiove automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases AT jussitohka automatedjointskullstrippingandsegmentationwithmultitaskunetinlargemousebrainmridatabases |