A deep learning framework for automated quantification of peripheral nerve lesions in MR neurography
Nedim C Beste, Clara Raudonat, Melissa Fesselier, Sebastian Hartmann, Daniel Sommer, Johann M E Jende, Jonathan Kottlors, Sabine Heiland, Hagen Meredig, Martin Bendszus, Christoph M Mooshage · 2026
Read A deep learning framework for automated quantification of peripheral nerve lesions in MR neurography by Nedim C Beste, Clara Raudonat, Melissa Fesselier, Sebastian Hartmann, Daniel Sommer, Johann M E Jende, Jonathan Kottlors, Sabine Heiland, Hagen Meredig, Martin Bendszus, Christoph M Mooshage (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-13553995.txt.
Description
Eleven authors build a deep learning system to measure nerve damage on MRI scans in this 2026 study. MR neurography gives sharp views of peripheral nerves, but counting the lesions inside a nerve by hand is slow and varies by reader.
The team trained a modified nnU-Net on scans of 178 patients, with nerves and lesions marked by hand. Nerve outlines were near perfect by the standard overlap score. Lesion outlines scored lower, as expected for tiny targets, but held up on two fresh test sets. The system also linked lesion burden to electrical nerve tests, a step toward fast, uniform measures for clinics and trials.
Catalog record
| CREATOR | Nedim C Beste, Clara Raudonat, Melissa Fesselier, Sebastian Hartmann, Daniel Sommer, Johann M E Jende, Jonathan Kottlors, Sabine Heiland, Hagen Meredig, Martin Bendszus, Christoph M Mooshage |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |