Few-shot learning to identify atypical endometrial hyperplasia and endometrial cancer based on transvaginal ultrasonic images
The proposed FSL model combining dual pretrained ResNet50 V2 eigenvectors extractor and KNN classifier presented well in identifying NAEH, AEH, and EC patients with limited TVU images, showing potential in the application of computer-aided disease diagnosis.
Authors
- Yindi Zhu (ORCID: https://orcid.org/0000-0003-3481-9578)
- Jinhua Zhou (ORCID: https://orcid.org/0000-0002-1163-002X)
- Youguo Chen (ORCID: https://orcid.org/0009-0003-1053-5367)
- Jin Ding
- Wen Liu (ORCID: https://orcid.org/0000-0002-8415-2161)
- Mingyue Wang (ORCID: https://orcid.org/0000-0001-8317-3414)
- Xinxian Gu
- Cui Feng
- Jinyan Bian
Institutions
- Soochow University (CN)
- Second Affiliated Hospital of Soochow University (CN)
- Suzhou Traditional Chinese Medicine Hospital (CN)
- Hangzhou Hospital of Traditional Chinese Medicine (CN)
- First Affiliated Hospital of Soochow University (CN)
- The Fourth Affiliated Hospital of Soochow University (Suzhou Dushu Lake Hospital)
Publication Details
- Journal
- Heliyon
- Published
- 2024-08-01
- DOI
- https://doi.org/10.1016/j.heliyon.2024.e36426
- Citations
- 6
- Primary Topic
- Endometrial and Cervical Cancer Treatments
- Type
- article
- Field-Weighted Citation Impact
- 3.50
Funders
- Clinical Special Fund of Jiangsu Province
- Government of Jiangsu Province
- Natick Soldier Research, Development and Engineering Center