Improved Segmentation of Deep Sulci in Cortical Gray Matter Using a Deep Learning Framework Incorporating Laplace’s Equation
(2023) 28th International Conference on Information Processing in Medical Imaging, IPMI 2023 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 13939 LNCS. p.692-704- Abstract
When developing tools for automated cortical segmentation, the ability to produce topologically correct segmentations is important in order to compute geometrically valid morphometry measures. In practice, accurate cortical segmentation is challenged by image artifacts and the highly convoluted anatomy of the cortex itself. To address this, we propose a novel deep learning-based cortical segmentation method in which prior knowledge about the geometry of the cortex is incorporated into the network during the training process. We design a loss function which uses the theory of Laplace’s equation applied to the cortex to locally penalize unresolved boundaries between tightly folded sulci. Using an ex vivo MRI dataset of human medial... (More)
When developing tools for automated cortical segmentation, the ability to produce topologically correct segmentations is important in order to compute geometrically valid morphometry measures. In practice, accurate cortical segmentation is challenged by image artifacts and the highly convoluted anatomy of the cortex itself. To address this, we propose a novel deep learning-based cortical segmentation method in which prior knowledge about the geometry of the cortex is incorporated into the network during the training process. We design a loss function which uses the theory of Laplace’s equation applied to the cortex to locally penalize unresolved boundaries between tightly folded sulci. Using an ex vivo MRI dataset of human medial temporal lobe specimens, we demonstrate that our approach outperforms baseline segmentation networks, both quantitatively and qualitatively.
(Less)
- author
- organization
- publishing date
- 2023
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Cortical segmentation, ex vivo MRI, topology correction
- host publication
- Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- editor
- Frangi, Alejandro ; de Bruijne, Marleen ; Wassermann, Demian and Navab, Nassir
- volume
- 13939 LNCS
- pages
- 13 pages
- publisher
- Springer Science and Business Media B.V.
- conference name
- 28th International Conference on Information Processing in Medical Imaging, IPMI 2023
- conference location
- San Carlos de Bariloche, Argentina
- conference dates
- 2023-06-18 - 2023-06-23
- external identifiers
-
- scopus:85163960679
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 9783031340475
- DOI
- 10.1007/978-3-031-34048-2_53
- language
- English
- LU publication?
- yes
- id
- ef96e585-b951-46ef-a794-6893641feb7a
- date added to LUP
- 2023-10-16 13:01:50
- date last changed
- 2024-04-19 02:22:24
@inproceedings{ef96e585-b951-46ef-a794-6893641feb7a, abstract = {{<p>When developing tools for automated cortical segmentation, the ability to produce topologically correct segmentations is important in order to compute geometrically valid morphometry measures. In practice, accurate cortical segmentation is challenged by image artifacts and the highly convoluted anatomy of the cortex itself. To address this, we propose a novel deep learning-based cortical segmentation method in which prior knowledge about the geometry of the cortex is incorporated into the network during the training process. We design a loss function which uses the theory of Laplace’s equation applied to the cortex to locally penalize unresolved boundaries between tightly folded sulci. Using an ex vivo MRI dataset of human medial temporal lobe specimens, we demonstrate that our approach outperforms baseline segmentation networks, both quantitatively and qualitatively.</p>}}, author = {{Ravikumar, Sadhana and Ittyerah, Ranjit and Lim, Sydney and Xie, Long and Das, Sandhitsu and Khandelwal, Pulkit and Wisse, Laura E.M. and Bedard, Madigan L. and Robinson, John L. and Schuck, Terry and Grossman, Murray and Trojanowski, John Q. and Lee, Edward B. and Tisdall, M. Dylan and Prabhakaran, Karthik and Detre, John A. and Irwin, David J. and Trotman, Winifred and Mizsei, Gabor and Artacho-Pérula, Emilio and de Onzono Martin, Maria Mercedes Iñiguez and del Mar Arroyo Jiménez, Maria and Muñoz, Monica and Romero, Francisco Javier Molina and del Pilar Marcos Rabal, Maria and Cebada-Sánchez, Sandra and González, José Carlos Delgado and de la Rosa-Prieto, Carlos and Parada, Marta Córcoles and Wolk, David A. and Insausti, Ricardo and Yushkevich, Paul A.}}, booktitle = {{Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings}}, editor = {{Frangi, Alejandro and de Bruijne, Marleen and Wassermann, Demian and Navab, Nassir}}, isbn = {{9783031340475}}, issn = {{1611-3349}}, keywords = {{Cortical segmentation; ex vivo MRI; topology correction}}, language = {{eng}}, pages = {{692--704}}, publisher = {{Springer Science and Business Media B.V.}}, series = {{Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}}, title = {{Improved Segmentation of Deep Sulci in Cortical Gray Matter Using a Deep Learning Framework Incorporating Laplace’s Equation}}, url = {{http://dx.doi.org/10.1007/978-3-031-34048-2_53}}, doi = {{10.1007/978-3-031-34048-2_53}}, volume = {{13939 LNCS}}, year = {{2023}}, }