Answers are grounded in this paper and cite their sources.
Introduction Cervical central canal (CCC) stenosis is associated with low CCC area. Traditional methods involve manual measurements, which could potentially be replaced by artificial intelligent (AI). This study evaluates the predictive value of shape approximations, specifically elliptical, triangular, and rectangular as well as second-order approximations (SOA), in estimating the CCC area using anteroposterior diameter (APD) and interpedicular distance (IPD) in patients with CCC stenosis. Research Question How accurately can shape approximations, including SOA, predict the CCC area in patients with stenosis? Material and Methods This retrospective cohort study analyzed cervical spine imaging data from 334 patients. Central canal measurements, including APD, IPD, and CCC area, were obtained from CT scans. The APD and IPD values were applied to pre-existing approximation equations to estimate the CCC area. These approximated areas were compared to manual measurements, with statistical analyses such as mean differences, percent error, and paired sample t-tests used to assess the accuracy of the approximations. Results Out of 334 patients, 118 met the inclusion criteria, with 239 stenotic levels assessed. The ellipse approximation showed mean differences from -2.95 mm² at C5-6 to -36.03 mm² at C3-4. It demonstrated no significant difference from manual measurements at C4-5, C5-6, and C6-7. SOAs outperformed single-shape approximations, with mean differences ranging from -0.46 mm² at C6-7 to -16.26 mm² at C2-3, and showed no significant difference at C3-4, C4-5, C5-6, and C6-7. Discussion and Conclusion Elliptical and SOA methods effectively predict CCC area in stenosis, demonstrating potential for integration into AI-based diagnostic systems.
No takes yet. Share an insight, caveat, or question.
Brandt et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: