Deep Learning outperforms physicians in myopathy and neuropathy classification based on Needle Electromyography Signal
Ilhan Yoo, Jaesung Yoo, Dongmin Kim, Ina Youn, Hyodong Kim, Michelle Youn, Jun Hee Won, Woosup Cho · 2026
Read Deep Learning outperforms physicians in myopathy and neuropathy classification based on Needle Electromyography Signal by Ilhan Yoo, Jaesung Yoo, Dongmin Kim, Ina Youn, Hyodong Kim, Michelle Youn, Jun Hee Won, Woosup Cho (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-0339691.txt.
Description
Ilhan Yoo and colleagues tested a deep learning system that reads needle electromyography signals, the electrical traces from muscles used to diagnose nerve and muscle disease. Human reading of these traces is slow and swayed by bias. Using three hundred seventy-six signals from fifty-seven patients at a university hospital, the model sorted patients into healthy, muscle disease, or nerve disease.
The model beat six experienced electromyographers on accuracy and every other score, reaching about seventy percent patient-level accuracy against the physicians' mid-fifties. Inspection showed it leaned on genuinely relevant signal features. One caution stood out: when the model was wrong, it was wrong in unison more often than the doctors, whose errors scattered. The study argues for AI as a second reader, not a replacement.
Catalog record
| CREATOR | Ilhan Yoo, Jaesung Yoo, Dongmin Kim, Ina Youn, Hyodong Kim, Michelle Youn, Jun Hee Won, Woosup Cho |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28334 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |