Revisiting Data Augmentation for Ultrasound Images
Adam Tupper [email protected], Christian Gagné [email protected] · 2025
Read Revisiting Data Augmentation for Ultrasound Images by Adam Tupper [email protected], Christian Gagné [email protected] (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/arxiv-2501-13193.txt.
Description
Adam Tupper and Christian Gagné revisit data augmentation for ultrasound images in two thousand twenty-five. Augmentation, making varied copies of training images, reliably helps deep networks generalize. Medical imaging often lacks data, yet augmentation is underused there because which tricks work for which scan type is unclear.
The authors built a standard benchmark of fourteen ultrasound tasks from ten sources, spanning eleven body regions and both classification and segmentation. They then measured many augmentation styles head to head. Tricks borrowed from ordinary photo tasks often beat ones designed for ultrasound. A diverse random scheme popular for photos also worked well here. The paper offers its testing method as a template for other scan types and settings.
Catalog record
| CREATOR | Adam Tupper [email protected], Christian Gagné [email protected] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 69 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-13193