Creation of a Catalogue of Adult Cervical Spine Morphometrics from Automatic Three-Dimensional Measurements of CT-Scan Images and Cluster Analysis
Cervical spine models used in biomechanical simulations often fail to represent an intended population by trimming it down to one average or specific subject. Better population representation could be achieved by creating a catalogue of archetypical spine morphologies. A total of 450 subjects were gathered from publicly available datasets of CT-Scans, originating from the United States, Canada, Switzerland and Germany. Vertebral templates with identified anatomical landmarks were created from a 50th-centile subject and were elastically registered to the subjects’ vertebrae. Based on the coordinates of the registered landmarks, anatomical measurements were taken (vertebral body height, width and depth; spinal canal width and depth; pedicle width, height and transverse angle; intervertebral disc height). Data analysis consisted of a hierarchical agglomerative clustering on principle components. The morphology of the cervical spine was clustered into four archetypes based on two uncorrelated dimensions (overall vertebral size, overall spinal canal size). Males presented bigger vertebrae than females ( p = 6 × 10 −74 ), and subjects older than 65 years old can present slightly bigger vertebrae ( p = 3 × 10 −2 ) and thinner intervertebral discs ( p = 8 × 10 −3 ). Subjects of any sex and age can present an inherently narrower or wider spinal canal. Measurements for cervical spine archetypes were provided and can be used as a morphing target by institutions willing to enhance the generalisation of their results. An inherently narrower spinal canal could have clinical implications for the elderly when degenerative stenosis progresses, and the proposed method of automatic measurement can be extended to any organ.
Authors
- Yvan Petit (ORCID: https://orcid.org/0000-0003-1428-8191)
- Julien Clin (ORCID: https://orcid.org/0000-0001-9545-5857)
- Jean‐Marc Mac‐Thiong (ORCID: https://orcid.org/0000-0003-4524-3893)
- Bahe Hachem (ORCID: https://orcid.org/0009-0009-0843-6873)
- Lucien Diotalevi (ORCID: https://orcid.org/0000-0001-9822-9951)
- Franck LeNaveaux (ORCID: https://orcid.org/0009-0007-5692-0825)
Institutions
- Centre Intégré Universitaire de Santé et de Services Sociaux du Centre-Sud-de-l'Île-de-Montréal (CA)
- Centres Intégré Universitaires de Santé et de Services Sociaux (CA)
- Université de Montréal (CA)
- École de Technologie Supérieure (CA)
Publication Details
- Journal
- Annals of Biomedical Engineering
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1007/s10439-026-04446-6
- Primary Topic
- Medical Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00