TY - JOUR
T1 - Automatic generation of subject-specific finite element models of the spine from magnetic resonance images
AU - Kok, Joeri
AU - Shcherbakova, Yulia M.
AU - Schlösser, Tom P.C.
AU - Seevinck, Peter R.
AU - van der Velden, Tijl A.
AU - Castelein, René M.
AU - Ito, Keita
AU - van Rietbergen, Bert
N1 - Funding Information:
The authors would like to thank the European Research Council (Grant no: 101020004) for providing financial support to this project.
Publisher Copyright:
Copyright © 2023 Kok, Shcherbakova, Schlösser, Seevinck, van der Velden, Castelein, Ito and van Rietbergen.
PY - 2023/9/5
Y1 - 2023/9/5
N2 - The generation of subject-specific finite element models of the spine is generally a time-consuming process based on computed tomography (CT) images, where scanning exposes subjects to harmful radiation. In this study, a method is presented for the automatic generation of spine finite element models using images from a single magnetic resonance (MR) sequence. The thoracic and lumbar spine of eight adult volunteers was imaged using a 3D multi-echo-gradient-echo sagittal MR sequence. A deep-learning method was used to generate synthetic CT images from the MR images. A pre-trained deep-learning network was used for the automatic segmentation of vertebrae from the synthetic CT images. Another deep-learning network was trained for the automatic segmentation of intervertebral discs from the MR images. The automatic segmentations were validated against manual segmentations for two subjects, one with scoliosis, and another with a spine implant. A template mesh of the spine was registered to the segmentations in three steps using a Bayesian coherent point drift algorithm. First, rigid registration was applied on the complete spine. Second, non-rigid registration was used for the individual discs and vertebrae. Third, the complete spine was non-rigidly registered to the individually registered discs and vertebrae. Comparison of the automatic and manual segmentations led to dice-scores of 0.93–0.96 for all vertebrae and discs. The lowest dice-score was in the disc at the height of the implant where artifacts led to under-segmentation. The mean distance between the morphed meshes and the segmentations was below 1 mm. In conclusion, the presented method can be used to automatically generate accurate subject-specific spine models.
AB - The generation of subject-specific finite element models of the spine is generally a time-consuming process based on computed tomography (CT) images, where scanning exposes subjects to harmful radiation. In this study, a method is presented for the automatic generation of spine finite element models using images from a single magnetic resonance (MR) sequence. The thoracic and lumbar spine of eight adult volunteers was imaged using a 3D multi-echo-gradient-echo sagittal MR sequence. A deep-learning method was used to generate synthetic CT images from the MR images. A pre-trained deep-learning network was used for the automatic segmentation of vertebrae from the synthetic CT images. Another deep-learning network was trained for the automatic segmentation of intervertebral discs from the MR images. The automatic segmentations were validated against manual segmentations for two subjects, one with scoliosis, and another with a spine implant. A template mesh of the spine was registered to the segmentations in three steps using a Bayesian coherent point drift algorithm. First, rigid registration was applied on the complete spine. Second, non-rigid registration was used for the individual discs and vertebrae. Third, the complete spine was non-rigidly registered to the individually registered discs and vertebrae. Comparison of the automatic and manual segmentations led to dice-scores of 0.93–0.96 for all vertebrae and discs. The lowest dice-score was in the disc at the height of the implant where artifacts led to under-segmentation. The mean distance between the morphed meshes and the segmentations was below 1 mm. In conclusion, the presented method can be used to automatically generate accurate subject-specific spine models.
KW - deep-learning
KW - intervertebral disc
KW - mesh morphing
KW - personalized medicine
KW - synthetic computed tomography
KW - vertebra
UR - http://www.scopus.com/inward/record.url?scp=85171529259&partnerID=8YFLogxK
U2 - 10.3389/fbioe.2023.1244291
DO - 10.3389/fbioe.2023.1244291
M3 - Article
AN - SCOPUS:85171529259
SN - 2296-4185
VL - 11
JO - Frontiers in bioengineering and biotechnology
JF - Frontiers in bioengineering and biotechnology
M1 - 1244291
ER -