Evaluation of an AI-based segmentation model for mandibular fracture segmentation on CBCT using volumetric and 2D slice-based validation approaches
Ilgın Arı, Ersagun Kara, Arzum Yılmaz, Ecem Usman, Alper Aktaş · 2026
Read Evaluation of an AI-based segmentation model for mandibular fracture segmentation on CBCT using volumetric and 2D slice-based validation approaches by Ilgın Arı, Ersagun Kara, Arzum Yılmaz, Ecem Usman, Alper Aktaş (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13499960.txt.
Description
Ilgın Arı and co-authors test in 2026 an artificial intelligence model that outlines jaw fractures on cone-beam CT scans, a three-dimensional dental imaging method. Accurate fracture mapping guides surgery for lower jaw breaks, a common facial injury. Scan data from 92 patients covering 142 fracture sites were gathered retrospectively, with two maxillofacial surgeons hand-drawing the reference outlines. A standard self-configuring network was trained on most cases and tested on the rest.
Performance was scored two ways, by whole volume and slice by slice, using overlap and distance measures. Volume-based scores beat slice-based scores on every reported metric, with moderate overlap accuracy overall and smaller surface errors in three dimensions. The authors conclude that judging such tools volumetrically gives a fuller and more reliable picture, even though absolute segmentation accuracy remains limited and needs improvement before clinical trust.
Catalog record
| CREATOR | Ilgın Arı, Ersagun Kara, Arzum Yılmaz, Ecem Usman, Alper Aktaş |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39416 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |