A comparative study of single-stage and dual-stage classification models for OPMDs
Jiayin Yu, Rui Huang, Xuan Wang, Yiran Zhang, Yueling Li, Yuhang Ye, Mei Ouyang, Weilan Su, Jing Wen, Zihe Zhang, Qinwen Liu, Zhiwen Liu, Wei Liu · 2025
Read A comparative study of single-stage and dual-stage classification models for OPMDs by Jiayin Yu, Rui Huang, Xuan Wang, Yiran Zhang, Yueling Li, Yuhang Ye, Mei Ouyang, Weilan Su, Jing Wen, Zihe Zhang, Qinwen Liu, Zhiwen Liu, Wei Liu (2025) 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-12836996.txt.
Description
Jiayin Yu and co-authors compare two ways of teaching computers to read mouth photographs for early disease, and release the photo set they built, in a 2025 study. Oral cancers are often caught late; disorders that precede them are visible but under-photographed in public data. Their new set holds one thousand three hundred forty-eight images across five classes, from healthy tissue to oral cancer. Ten standard image models were trained two ways.
In the single-stage route, one model sorts all five classes at once. In the dual-stage route, a first model splits broad groups and a second refines them. The best dual-stage pairing reached about ninety percent accuracy, beating the best single-stage model's eighty-seven. The authors release the data set openly and pitch the two-stage design for clinics screening mouth lesions with a camera and a model.
Catalog record
| CREATOR | Jiayin Yu, Rui Huang, Xuan Wang, Yiran Zhang, Yueling Li, Yuhang Ye, Mei Ouyang, Weilan Su, Jing Wen, Zihe Zhang, Qinwen Liu, Zhiwen Liu, Wei Liu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |