Geometric learning and statistical modeling for surgical outcomes evaluation in craniosynostosis using 3D photogrammetry.
3D photogrammetry
Craniofacial imaging
Craniosynostosis
Geometric learning
Graph convolutional neural network
Landmark detection
Journal
Computer methods and programs in biomedicine
ISSN: 1872-7565
Titre abrégé: Comput Methods Programs Biomed
Pays: Ireland
ID NLM: 8506513
Informations de publication
Date de publication:
Oct 2023
Oct 2023
Historique:
received:
18
01
2023
revised:
11
05
2023
accepted:
22
06
2023
pmc-release:
01
10
2024
medline:
29
8
2023
pubmed:
3
7
2023
entrez:
2
7
2023
Statut:
ppublish
Résumé
Accurate and repeatable detection of craniofacial landmarks is crucial for automated quantitative evaluation of head development anomalies. Since traditional imaging modalities are discouraged in pediatric patients, 3D photogrammetry has emerged as a popular and safe imaging alternative to evaluate craniofacial anomalies. However, traditional image analysis methods are not designed to operate on unstructured image data representations such as 3D photogrammetry. We present a fully automated pipeline to identify craniofacial landmarks in real time, and we use it to assess the head shape of patients with craniosynostosis using 3D photogrammetry. To detect craniofacial landmarks, we propose a novel geometric convolutional neural network based on Chebyshev polynomials to exploit the point connectivity information in 3D photogrammetry and quantify multi-resolution spatial features. We propose a landmark-specific trainable scheme that aggregates the multi-resolution geometric and texture features quantified at every vertex of a 3D photogram. Then, we embed a new probabilistic distance regressor module that leverages the integrated features at every point to predict landmark locations without assuming correspondences with specific vertices in the original 3D photogram. Finally, we use the detected landmarks to segment the calvaria from the 3D photograms of children with craniosynostosis, and we derive a new statistical index of head shape anomaly to quantify head shape improvements after surgical treatment. We achieved an average error of 2.74 ± 2.70 mm identifying Bookstein Type I craniofacial landmarks, which is a significant improvement compared to other state-of-the-art methods. Our experiments also demonstrated a high robustness to spatial resolution variability in the 3D photograms. Finally, our head shape anomaly index quantified a significant reduction of head shape anomalies as a consequence of surgical treatment. Our fully automated framework provides real-time craniofacial landmark detection from 3D photogrammetry with state-of-the-art accuracy. In addition, our new head shape anomaly index can quantify significant head phenotype changes and can be used to quantitatively evaluate surgical treatment in patients with craniosynostosis.
Sections du résumé
BACKGROUND AND OBJECTIVE
OBJECTIVE
Accurate and repeatable detection of craniofacial landmarks is crucial for automated quantitative evaluation of head development anomalies. Since traditional imaging modalities are discouraged in pediatric patients, 3D photogrammetry has emerged as a popular and safe imaging alternative to evaluate craniofacial anomalies. However, traditional image analysis methods are not designed to operate on unstructured image data representations such as 3D photogrammetry.
METHODS
METHODS
We present a fully automated pipeline to identify craniofacial landmarks in real time, and we use it to assess the head shape of patients with craniosynostosis using 3D photogrammetry. To detect craniofacial landmarks, we propose a novel geometric convolutional neural network based on Chebyshev polynomials to exploit the point connectivity information in 3D photogrammetry and quantify multi-resolution spatial features. We propose a landmark-specific trainable scheme that aggregates the multi-resolution geometric and texture features quantified at every vertex of a 3D photogram. Then, we embed a new probabilistic distance regressor module that leverages the integrated features at every point to predict landmark locations without assuming correspondences with specific vertices in the original 3D photogram. Finally, we use the detected landmarks to segment the calvaria from the 3D photograms of children with craniosynostosis, and we derive a new statistical index of head shape anomaly to quantify head shape improvements after surgical treatment.
RESULTS
RESULTS
We achieved an average error of 2.74 ± 2.70 mm identifying Bookstein Type I craniofacial landmarks, which is a significant improvement compared to other state-of-the-art methods. Our experiments also demonstrated a high robustness to spatial resolution variability in the 3D photograms. Finally, our head shape anomaly index quantified a significant reduction of head shape anomalies as a consequence of surgical treatment.
CONCLUSION
CONCLUSIONS
Our fully automated framework provides real-time craniofacial landmark detection from 3D photogrammetry with state-of-the-art accuracy. In addition, our new head shape anomaly index can quantify significant head phenotype changes and can be used to quantitatively evaluate surgical treatment in patients with craniosynostosis.
Identifiants
pubmed: 37393741
pii: S0169-2607(23)00354-1
doi: 10.1016/j.cmpb.2023.107689
pmc: PMC10527531
mid: NIHMS1914145
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
107689Subventions
Organisme : NIDCR NIH HHS
ID : R00 DE027993
Pays : United States
Organisme : NCATS NIH HHS
ID : TL1 TR002533
Pays : United States
Informations de copyright
Copyright © 2023. Published by Elsevier B.V.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors have no competing financial interests to declare.
Références
Plast Reconstr Surg. 2017 Jul;140(1):138-145
pubmed: 28654600
Am J Med Genet A. 2021 Sep;185(9):2670-2675
pubmed: 34008890
Plast Reconstr Surg Glob Open. 2022 Aug 10;10(8):e4457
pubmed: 35983543
J Craniofac Surg. 2021 May 1;32(3):956-963
pubmed: 33405445
Plast Reconstr Surg. 2019 Dec;144(6):1051e-1060e
pubmed: 31764657
Int J Comput Assist Radiol Surg. 2021 Feb;16(2):277-287
pubmed: 33417161
IEEE Trans Pattern Anal Mach Intell. 2023 Jun;45(6):6923-6939
pubmed: 33872143
Diagnostics (Basel). 2022 Jun 21;12(7):
pubmed: 35885422
Plast Reconstr Surg. 2017 Oct;140(4):579e-586e
pubmed: 28617738
J Plast Reconstr Aesthet Surg. 2021 Mar;74(3):605-614
pubmed: 33082078
Plast Reconstr Surg. 2018 Nov;142(5):708e-717e
pubmed: 30113443
Med Image Anal. 2014 May;18(4):635-46
pubmed: 24713202
J Craniofac Surg. 2023 Jul-Aug 01;34(5):1485-1488
pubmed: 36944601
Childs Nerv Syst. 2021 Feb;37(2):573-579
pubmed: 32812117
BMC Bioinformatics. 2013 Jul 22;14:232
pubmed: 23870191
IEEE Trans Med Imaging. 2021 Dec;40(12):3867-3878
pubmed: 34310293
Sci Rep. 2020 Mar 10;10(1):4443
pubmed: 32157192
J Craniomaxillofac Surg. 2015 Nov;43(9):1798-803
pubmed: 26421470
Phys Med Biol. 2023 Apr 26;68(9):
pubmed: 36652722
J Craniofac Surg. 2020 May/Jun;31(3):697-701
pubmed: 32011542
J Craniofac Surg. 2021 Jan-Feb 01;32(1):371-450
pubmed: 33156164
IEEE Trans Biomed Eng. 2006 Aug;53(8):1615-23
pubmed: 16916096
Sci Rep. 2020 Sep 18;10(1):15346
pubmed: 32948813
Cleft Palate Craniofac J. 2011 Jul;48(4):394-8
pubmed: 20815706
J Neurosurg Pediatr. 2016 Sep;18(3):275-80
pubmed: 27153376
Head Face Med. 2010 Jul 28;6:18
pubmed: 20667081
J Craniofac Surg. 2012 May;23(3):799-804
pubmed: 22565928
Cleft Palate Craniofac J. 2023 Aug;60(8):971-979
pubmed: 35306870
IEEE J Biomed Health Inform. 2021 Jul;25(7):2643-2654
pubmed: 33147152
Plast Reconstr Surg. 2011 Jun;127(6):2397-2406
pubmed: 21311385
Sensors (Basel). 2020 Jun 03;20(11):
pubmed: 32503190
IEEE Trans Med Imaging. 2019 Apr;38(4):919-931
pubmed: 30334750
IEEE Trans Biomed Eng. 2022 Feb;69(2):537-546
pubmed: 34324420
Plast Reconstr Surg. 2006 Feb;117(2):532-41
pubmed: 16462336