Validation of deep learning enabled web based and smartphone optimized application RadAnalyzer to measure vertebral heart size and vertebral left atrial size in dogs
Sonya Gordon, Tomas Reyes, Tabitha Baibos-Reyes, Katharine Tess Sykes, Sukjung Lim, Alice Watson · 2026
Read Validation of deep learning enabled web based and smartphone optimized application RadAnalyzer to measure vertebral heart size and vertebral left atrial size in dogs by Sonya Gordon, Tomas Reyes, Tabitha Baibos-Reyes, Katharine Tess Sykes, Sukjung Lim, Alice Watson (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/plos-pone-0337679.txt.
Description
Sonya Gordon and colleagues validated an artificial intelligence tool that measures dog heart size on chest X-rays, a routine check whose hand measurements vary between observers. The tool, RadAnalyzer, runs on the web and on phones and reports two standard scores comparing heart dimensions against backbone length. Its readings on more than a thousand dogs of eighty breeds were set against a trained human observer's.
Agreement was strong for both scores, with average differences amounting to almost nothing and scatter well inside what clinicians tolerate. The authors say the tool can safely assist veterinarians measuring good-quality side-view films in dogs of any size. They note the next test is against ultrasound heart measurement, the deeper standard, before the scores can stand in for fuller cardiac workups.
Catalog record
| CREATOR | Sonya Gordon, Tomas Reyes, Tabitha Baibos-Reyes, Katharine Tess Sykes, Sukjung Lim, Alice Watson |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 19747 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |