Automated Lesion Segmentation in Medical Imaging via Integration of nnU-Net Optimization and SAM Approach
Alejandro Jerónimo, Ignacio Rojas, Olga Valenzuela · 2026
Read Automated Lesion Segmentation in Medical Imaging via Integration of nnU-Net Optimization and SAM Approach by Alejandro Jerónimo, Ignacio Rojas, Olga Valenzuela (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-13157531.txt.
Description
Alejandro Jerónimo, Ignacio Rojas, and Olga Valenzuela join two well-known tools into one automatic pipeline for outlining lesions in medical scans. Their 2026 paper pairs nnU-Net, which tunes itself for a dataset, with the Segment Anything Model, a broad image tool that usually needs a person to point at the target. The goal is sharp outlines with no hand guidance.
The team tested the pipeline on lung nodules in the LIDC-IDRI dataset, a standard public benchmark for this task. The first tool finds each nodule, and the second then draws its border. The joined method produced outlines that followed the true shape of the nodules more closely than the first tool alone. In many cases the borders looked truer than single expert drawings, which vary a lot from one reader to the next.
Catalog record
| CREATOR | Alejandro Jerónimo, Ignacio Rojas, Olga Valenzuela |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |