CaliDent-Net: domain-constrained self-supervised pre-training with parallel attention and prototype calibration for dental radiograph analysis
Yanyu Miao, Xiaoqin Zhang, Fei Leng, Yanling Zhu · 2026
Read CaliDent-Net: domain-constrained self-supervised pre-training with parallel attention and prototype calibration for dental radiograph analysis by Yanyu Miao, Xiaoqin Zhang, Fei Leng, Yanling Zhu (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/pmc-13576453.txt.
Description
Yanyu Miao, Xiaoqin Zhang, Fei Leng, and Yanling Zhu build a neural network that reads dental X-rays with well-judged confidence, in a paper dated twenty twenty-six. Dental X-ray sets with expert labels are small, and standard models often sound sure when they are wrong. Their system pretrains itself on unlabeled films using changes that respect tooth and bone shape. An attention block then weighs space and channel clues at once.
A final prototype step classes each film by likeness to stored examples, which also gives a case-based reason for the call. On a public set of 1,272 images across four disease classes, it reached about ninety-six percent accuracy. Scores beat eight rival models, and trust in its confidence improved.
Catalog record
| CREATOR | Yanyu Miao, Xiaoqin Zhang, Fei Leng, Yanling Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |