TVnet : Automated Time-Resolved Tracking of the Tricuspid Valve Plane in MRI Long-Axis Cine Images with a Dual-Stage Deep Learning Pipeline
(2021) 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 12906 LNCS. p.567-576- Abstract
Tracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-demanding as the TV moves rapidly and is barely distinguishable from the myocardium. This study presents TVnet, a novel dual-stage deep learning pipeline based on ResNet-50 and automated image linear transformation, able to automatically derive tricuspid annular plane systolic excursion. Stage 1 uses a trained network for a coarse detection of the TV points, which are used by stage 2 to reorient the cine into a standardized size, cropping,... (More)
Tracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-demanding as the TV moves rapidly and is barely distinguishable from the myocardium. This study presents TVnet, a novel dual-stage deep learning pipeline based on ResNet-50 and automated image linear transformation, able to automatically derive tricuspid annular plane systolic excursion. Stage 1 uses a trained network for a coarse detection of the TV points, which are used by stage 2 to reorient the cine into a standardized size, cropping, resolution, and heart orientation and to accurately locate the TV points with another trained network. The model was trained and evaluated on 4170 images from 140 patients with diverse cardiovascular pathologies. A baseline model without standardization achieved a Euclidean distance error of 4.0 ± 3.1 mm and a clinical-metric agreement of ICC = 0.87, whereas a standardized model improved the agreement to 2.4 ± 1.7 mm and an ICC = 0.94, on par with an evaluated inter-observer variability of 2.9 ± 2.9 mm and an ICC = 0.92, respectively. This novel dual-stage deep learning pipeline substantially improved the annotation accuracy compared to a baseline model, paving the way towards reliable right ventricular dysfunction assessment with MRI.
(Less)
- author
- Gonzales, Ricardo A.
LU
; Lamy, Jérôme ; Seemann, Felicia LU ; Heiberg, Einar LU ; Onofrey, John A. and Peters, Dana C.
- organization
- publishing date
- 2021
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Annotation, Cine MRI, Residual neural networks, Right ventricular dysfunction
- host publication
- Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 - 24th International Conference, Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- editor
- de Bruijne, Marleen ; Cattin, Philippe C. ; Cotin, Stéphane ; Padoy, Nicolas ; Speidel, Stefanie ; Zheng, Yefeng and Essert, Caroline
- volume
- 12906 LNCS
- pages
- 10 pages
- publisher
- Springer Science and Business Media B.V.
- conference name
- 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
- conference location
- Virtual, Online
- conference dates
- 2021-09-27 - 2021-10-01
- external identifiers
-
- scopus:85116509480
- ISSN
- 0302-9743
- 1611-3349
- ISBN
- 9783030872304
- DOI
- 10.1007/978-3-030-87231-1_55
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2021, Springer Nature Switzerland AG.
- id
- 07904167-4f30-4484-b7f4-0e7b7e363a93
- date added to LUP
- 2021-10-25 14:16:08
- date last changed
- 2025-02-09 18:51:24
@inproceedings{07904167-4f30-4484-b7f4-0e7b7e363a93, abstract = {{<p>Tracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-demanding as the TV moves rapidly and is barely distinguishable from the myocardium. This study presents TVnet, a novel dual-stage deep learning pipeline based on ResNet-50 and automated image linear transformation, able to automatically derive tricuspid annular plane systolic excursion. Stage 1 uses a trained network for a coarse detection of the TV points, which are used by stage 2 to reorient the cine into a standardized size, cropping, resolution, and heart orientation and to accurately locate the TV points with another trained network. The model was trained and evaluated on 4170 images from 140 patients with diverse cardiovascular pathologies. A baseline model without standardization achieved a Euclidean distance error of 4.0 ± 3.1 mm and a clinical-metric agreement of ICC = 0.87, whereas a standardized model improved the agreement to 2.4 ± 1.7 mm and an ICC = 0.94, on par with an evaluated inter-observer variability of 2.9 ± 2.9 mm and an ICC = 0.92, respectively. This novel dual-stage deep learning pipeline substantially improved the annotation accuracy compared to a baseline model, paving the way towards reliable right ventricular dysfunction assessment with MRI.</p>}}, author = {{Gonzales, Ricardo A. and Lamy, Jérôme and Seemann, Felicia and Heiberg, Einar and Onofrey, John A. and Peters, Dana C.}}, booktitle = {{Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 - 24th International Conference, Proceedings}}, editor = {{de Bruijne, Marleen and Cattin, Philippe C. and Cotin, Stéphane and Padoy, Nicolas and Speidel, Stefanie and Zheng, Yefeng and Essert, Caroline}}, isbn = {{9783030872304}}, issn = {{0302-9743}}, keywords = {{Annotation; Cine MRI; Residual neural networks; Right ventricular dysfunction}}, language = {{eng}}, pages = {{567--576}}, 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 = {{TVnet : Automated Time-Resolved Tracking of the Tricuspid Valve Plane in MRI Long-Axis Cine Images with a Dual-Stage Deep Learning Pipeline}}, url = {{http://dx.doi.org/10.1007/978-3-030-87231-1_55}}, doi = {{10.1007/978-3-030-87231-1_55}}, volume = {{12906 LNCS}}, year = {{2021}}, }