Abstract
Synthesizing panoramic radiographs from cone-beam com- puted tomography (CBCT) eliminates redundant patient radiation expo- sure and, in the absence of any public paired CBCT-panoramic dataset, provides the only scalable path to training data generation for down- stream dental imaging tasks. The quality of any panoramic reconstruc- tion is fundamentally determined by how accurately the dental arch spline traces the patient-specific curvature of the jaw, yet standard ge- ometric approaches fail on the missing-teeth and implant-bearing cases that constitute a significant proportion of clinical CBCT cases. We present AutoSpline, an open-source platform integrating three complementary arch-detection strategies: an interactive geometric tool, an automatic thresholding-based pipeline, and a learned heatmap regression network with geometric baseline conditioning trained on 163 expert-annotated arch splines from ToothFairy2. All three feed a shared Curved Planar Reformation renderer, and an interactive GUI allows clinicians to in- spect, correct, and export splines in Fiducial CSV (FCSV) format. We evaluate spline accuracy using mean curve distance (MCD) across three clinical categories and panoramic reconstruction quality using SSIM and PSNR.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/ODIN_030.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: https://papers.miccai.org/miccai-2026-sat/supp/ODIN_030_supp.pdf
Link to Open Review
Open Review Page: https://openreview.net/forum?id=Ly7pfzFQ6q
BibTex
@InProceedings{HosGol_Automated_MICCAISAT2026,
author = { Hosseinimanesh, Golriz AND Chafi, Imane AND Ma, Coty AND Cheriet, Farida AND Keren, Julia AND Landry-Schonbeck, Anaïs AND Huynh, Nelly AND Guibault, François},
title = { { Automated Dental Arch Curve Detection from CBCT: A Geometric and Learning-Based Platform } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17270},
month = {pending},
page = {pending}
}