Reference : DBSegment: Fast and robust segmentation of deep brain structures considering domain g...
Scientific journals : Article
Engineering, computing & technology : Multidisciplinary, general & others
Computational Sciences
http://hdl.handle.net/10993/52935
DBSegment: Fast and robust segmentation of deep brain structures considering domain generalisation
English
Baniasadi, Mehri mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Systems Control >]
Petersen, Mikkel V. mailto [University of Aarhus]
Goncalves, Jorge mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Systems Control >]
Horn, Andreas mailto [Harvard University > Harvard Medical School > > ; Charite Medical University of Berlin]
Vlasov, Vanja mailto []
Hertel, Frank mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > >]
Husch, Andreas mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Interventional Neuroscience >]
17-Oct-2022
Human Brain Mapping
John Wiley & Sons
Yes
International
1065-9471
1097-0193
Hoboken
United States - New York
[en] Segmentation ; Deep Learning ; MRI ; Confounder
[en] Segmenting deep brain structures from magnetic resonance images is important for patient diagnosis, surgical planning, and research. Most current state-of-the-art solutions follow a segmentation-by-registration approach, where subject magnetic resonance imaging (MRIs) are mapped to a template with well-defined segmentations. However, registration-based pipelines are time-consuming, thus, limiting their clinical use. This paper uses deep learning to provide a one-step, robust, and efficient deep brain segmentation solution directly in the native space. The method consists of a preprocessing step to conform all MRI images to the same orientation, followed by a convolutional neural network using the nnU-Net framework. We use a total of 14 datasets from both research and clinical collections. Of these, seven were used for training and validation and seven were retained for testing. We trained the network to segment 30 deep brain structures, as well as a brain mask, using labels generated from a registration-based approach. We evaluated the generalizability of the network by performing a leave-one-dataset-out cross-validation, and independent testing on unseen datasets. Furthermore, we assessed cross-domain transportability by evaluating the results separately on different domains. We achieved an average dice score similarity of 0.89 ± 0.04 on the test datasets when compared to the registration-based gold standard. On our test system, the computation time decreased from 43 min for a reference registration-based pipeline to 1.3 min. Our proposed method is fast, robust, and generalizes with high reliability. It can be extended to the segmentation of other brain structures. It is publicly available on GitHub, and as a pip package for convenient usage.
Fonds National de la Recherche - FnR ; Lundbeckfonden ; Emmy Noether stipend ; jascha fonden ; deutsche Forschungsgemeinschaft ; Berlin institute ion health ; 111 Project on computational intelligence and intelligent control ; National institute of health
Researchers ; Professionals ; Students ; Others
http://hdl.handle.net/10993/52935
10.1002/hbm.26097
https://doi.org/10.1002/hbm.26097
FnR ; FNR12548237 > Mehri Baniasadi > TreCoDBS > Personalised Tremor Control By Advanced Clinical Deep Brain Stimulation. > 01/11/2018 > 31/10/2022 > 2018

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
Human Brain Mapping - 2022 - Baniasadi - DBSegment Fast and robust segmentation of deep brain structures considering.pdfPublisher postprint2.93 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.