S2098
Physics - Image acquisition and processing
ESTRO 2026
identify non-physical deformations. Registration accuracy was quantified using mean distance-to- agreement (mDTA).Correlations between registration accuracy and age were analysed using Spearman rank correlation.
Results: Hypothalamus segmentation failed for 5 patients (2 medulloblastoma, 3 ependymoma). Four non-physical deformations were detected: three HV (19M, 13M, 12M) and one medulloblastoma patient (17M) registered to the 1.75-2.25y reference.Across all registrations (n=455), mDTA varied between structures, with the cerebellum producing the largest median mDTA (medulloblastoma: 7.5 mm, ependymoma: 2.2 mm, HVs: 1.7 mm), likely related to tumour presence and surgical effects. All other structures had median mDTA<2.3mm. A weak but significant association was observed between age and mDTA (Spearman ρ =0.11, p=0.02). This relationship was most evident for HVs ( ρ =0.46, p<0.001). For medulloblastoma ( ρ =0.18, p=0.04) and ependymoma ( ρ =-0.06, p=0.45), this relationship may have been obscured by pathology mismatch.Median mean mDTA was greatest for the 1.75-2.25y reference (1.57±0.03mm). 4.5-8.5y (1.48±0.04mm) and 7-11y (1.49±0.07mm) were similar, however 7-11y showed greater variability (SD; 1.75-2.25y:0.71mm, 4.5- 8.5y:0.93mm, 7-11y:1.41mm).
Conclusion: Registration accuracy is influenced by the age mismatch between moving and reference images. The strength and direction of this effect vary across diagnostic groups, suggesting that disease-related anatomical variability compounds age-dependent registration issues. A mid-childhood (4.5-8.5y) population atlas provides robust SN across a broad paediatric age range and disease types but diagnosis- specific anatomical alterations, particularly in the cerebellum, remain a source of registration variability. References: [1] Green, A, et al. Front Oncol. 2020;10:1178. doi:10.3389/fonc.2020.01178.[2] Peterson, MR, et al. J Neuro. Pediatr. 2021;28(4):458–68. doi:10.3171/2021.2.PEDS201006.[3] Jernigan, TL, et al. Neuroimage. 2015;124:1149–54. 10.1016/j.neuroimage.2015.04.057[4] Henschel, L, et al. Neuroimage. 2020;219:117012. doi:10.1016/j.neuroimage.2020.117012.[5] Fonov, VS, et al. NeuroImage, Volume 47, Supplement 1, 2009, S102. doi.org/10.1016/S1053-8119(09)70884-5[6] Avants, BB, et al. Med Image Anal. 2008;12(1):26–41. doi:10.1016/j.media.2007.06.004. Keywords: Image-based data mining, paediatric, registration
Made with FlippingBook - Share PDF online