TY - GEN

T1 - Deep Implicit Statistical Shape Models for 3D Lumbar Vertebrae Image Delineation

AU - Ocepek, Domen

AU - Podobnik, Gašper

AU - Ibragimov, Bulat

AU - Vrtovec, Tomaž

N1 - Publisher Copyright: © 2024 SPIE.

PY - 2024

Y1 - 2024

N2 - Spinal imaging serves as an invaluable tool in the non-invasive visualization and evaluation of spinal pathologies. A key basis for quantitative medical image analysis pertinent to clinical diagnosis and spinal surgery planning is the segmentation of vertebrae in computed tomography (CT) images. While fully convolutional networks in general dominate over medical image segmentation, with the U-Net being the architecture of choice, alternative methodologies may offer potential advancements. One promising approach is the deep implicit statistical shape model (DISSM), known for generating high-quality surfaces without discretization and for its robustness, underpinned by the use of rich and explicit anatomical priors, particularly for challenging cross-dataset clinical samples. This paper explores the utilization of DISSM for vertebra segmentation on two image datasets: a collection of 1005 CT spine images known as CTSpine1K for the shape decoder, and a set of 319 CT images known as VerSe2020 for the pose estimation encoders (translation, rotation, scaling and principal component analysis). These images and their corresponding vertebra segmentations are used for the preparation, preprocessing, and training and testing of DISSM. The preprocessing and learning techniques are based on a DISSM software package (AshStuff/dissm) with our custom modifications. The obtained segmentation results yielded an overall mean Dice coefficient of 0.767, average symmetric surface distance of 1.93 mm, and 95th percentile Hausdorff distance of 5.71 mm. We can therefore conclude that DISSM has the potential to further advance the field of vertebra segmentation.

AB - Spinal imaging serves as an invaluable tool in the non-invasive visualization and evaluation of spinal pathologies. A key basis for quantitative medical image analysis pertinent to clinical diagnosis and spinal surgery planning is the segmentation of vertebrae in computed tomography (CT) images. While fully convolutional networks in general dominate over medical image segmentation, with the U-Net being the architecture of choice, alternative methodologies may offer potential advancements. One promising approach is the deep implicit statistical shape model (DISSM), known for generating high-quality surfaces without discretization and for its robustness, underpinned by the use of rich and explicit anatomical priors, particularly for challenging cross-dataset clinical samples. This paper explores the utilization of DISSM for vertebra segmentation on two image datasets: a collection of 1005 CT spine images known as CTSpine1K for the shape decoder, and a set of 319 CT images known as VerSe2020 for the pose estimation encoders (translation, rotation, scaling and principal component analysis). These images and their corresponding vertebra segmentations are used for the preparation, preprocessing, and training and testing of DISSM. The preprocessing and learning techniques are based on a DISSM software package (AshStuff/dissm) with our custom modifications. The obtained segmentation results yielded an overall mean Dice coefficient of 0.767, average symmetric surface distance of 1.93 mm, and 95th percentile Hausdorff distance of 5.71 mm. We can therefore conclude that DISSM has the potential to further advance the field of vertebra segmentation.

KW - computed tomography (CT)

KW - deep implicit statistical shape model (DISSM)

KW - Deep learning

KW - principal component analysis (PCA)

KW - vertebra segmentation

U2 - 10.1117/12.3007664

DO - 10.1117/12.3007664

M3 - Article in proceedings

AN - SCOPUS:85193521005

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2024

A2 - Colliot, Olivier

A2 - Mitra, Jhimli

PB - SPIE

T2 - Medical Imaging 2024: Image Processing

Y2 - 19 February 2024 through 22 February 2024

ER -