Language-supervised electrocardiogram interpretation with vision transformer
Xue Zhou, Tianhui Li, Keijiro Nakamura, Hiromasa Hayama, Shing-Hong Liu, Tetsuo Sasano, Wenxi Chen, Xin Zhu · 2026
Read Language-supervised electrocardiogram interpretation with vision transformer by Xue Zhou, Tianhui Li, Keijiro Nakamura, Hiromasa Hayama, Shing-Hong Liu, Tetsuo Sasano, Wenxi Chen, Xin Zhu (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/pmc-13661825.txt.
Description
Xue Zhou and co-authors teach a computer model to read heart tracings with words in a 2026 study. Most models for these tracings, called ECGs, need a fresh labeled set for each heart condition. That need limits how fast they can spread to new tasks. The team instead trained one model on tracings paired with written summary reports.
The model pairs a tracing reader with a text reader and learns to match them. It was trained on a vast set of tracing and report pairs from a hospital database. At test time, plain word prompts named each condition, with no retraining for the task. Results were strongest for rhythm problems and weaker for shape based conditions, and held up on an outside data set.
Catalog record
| CREATOR | Xue Zhou, Tianhui Li, Keijiro Nakamura, Hiromasa Hayama, Shing-Hong Liu, Tetsuo Sasano, Wenxi Chen, Xin Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |