LiteFreqMamba:lightweight frequency-enhanced Mamba for efficient 3D brain tumor segmentation

Xiangning Hou, Jun Yao, Qiaochu Li, Caocao Xu, Wenxin Dong · 2026

Read LiteFreqMamba:lightweight frequency-enhanced Mamba for efficient 3D brain tumor segmentation by Xiangning Hou, Jun Yao, Qiaochu Li, Caocao Xu, Wenxin Dong (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-13585600.txt.

Description

Xiangning Hou, Jun Yao, Qiaochu Li, Caocao Xu, and Wenxin Dong present a light model to outline brain tumors in 3D scans in a 2026 paper. State space models suit the task since they scale in a straight line with size. Such models tend to favor smooth, low detail signals and can blur tumor edges.

Their model adds parts that stress fine edge detail in the early layers. A hybrid block joins long range mixing with focused attention to keep the space map intact. Skip links are tuned to cut noise in the detail feed. Tests on two standard tumor sets show the model beats other light rivals on both speed and accuracy.

Catalog record

CREATORXiangning Hou, Jun Yao, Qiaochu Li, Caocao Xu, Wenxin Dong
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13585600

Browse the full library · About the Tower of Babel