LPC-SonoNet: A Lightweight Network Based on SonoNet and Light Pyramid Convolution for Fetal Ultrasound Standard Plane Detection
Tianxiang Yu, Po-Hsiang Tsui, Denis Leonov, Shuicai Wu, Guangyu Bin, Zhuhuang Zhou, Zhe-Ming Lu · 2024
Read LPC-SonoNet: A Lightweight Network Based on SonoNet and Light Pyramid Convolution for Fetal Ultrasound Standard Plane Detection by Tianxiang Yu, Po-Hsiang Tsui, Denis Leonov, Shuicai Wu, Guangyu Bin, Zhuhuang Zhou, Zhe-Ming Lu (2024) 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-11644059.txt.
Description
Tianxiang Yu and co-authors present a lighter neural network for fetal ultrasound checks in this 2024 paper. Finding the standard view planes in pregnancy scans is key to spotting birth defects early. A known network for the task works well but is large, which limits use on modest hardware.
The team added a light pyramid block, which reads features at several scales with few settings, and removed one block from the base network. Trained on 12,400 public scan images, the new network scored 97.0 percent accuracy on six view classes and 91.9 percent on nine classes, slightly above the base network. It used 4.3 million settings against 14.9 million, a much smaller size.
Catalog record
| CREATOR | Tianxiang Yu, Po-Hsiang Tsui, Denis Leonov, Shuicai Wu, Guangyu Bin, Zhuhuang Zhou, Zhe-Ming Lu |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |