Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Improved Segmentation of Deep Sulci in Cortical Gray Matter Using a Deep Learning Framework Incorporating Laplace’s Equation

Ravikumar, Sadhana ; Ittyerah, Ranjit ; Lim, Sydney ; Xie, Long ; Das, Sandhitsu ; Khandelwal, Pulkit ; Wisse, Laura E.M. LU orcid ; Bedard, Madigan L. ; Robinson, John L. and Schuck, Terry , et al. (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)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
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}},
}