Universal Lesion Segmentation Challenge 2023: A Comparative Research of Different Algorithms

Kaiwen Shi, Yifei Li, Binh Ho, Jovian Wang, Kobe Guo · 2025

Read Universal Lesion Segmentation Challenge 2023: A Comparative Research of Different Algorithms by Kaiwen Shi, Yifei Li, Binh Ho, Jovian Wang, Kobe Guo (2025) 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/arxiv-2502-10608.txt.

Description

Kaiwen Shi, Yifei Li, Binh Ho, Jovian Wang, and Kobe Guo compare methods for outlining lesions in medical scans, in a paper from 2025. Machine learning can trace lesions in many tissue types, but no single model yet works well on all of them. A shared contest in 2023 was set up to push toward a universal tool.

The team designed their own layouts, tried many known layouts, and compared speed and accuracy. They settled on a design that mixes windowed attention with a U-shaped network. The paper records what worked, what failed, and why, and it points to next steps worth trying for a faster universal model.

Catalog record

CREATORKaiwen Shi, Yifei Li, Binh Ho, Jovian Wang, Kobe Guo
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE16 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-10608

Browse the full library · About the Tower of Babel