A hybrid CNN-Transformer network integrating multiscale spatially detailed features for medical image segmentation

Bo Li, Wei Zhou, Haijun Li · 2026

Read A hybrid CNN-Transformer network integrating multiscale spatially detailed features for medical image segmentation by Bo Li, Wei Zhou, Haijun Li (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-0345549.txt.

Description

Bo Li, Wei Zhou, and Haijun Li built a hybrid AI model for medical scans. Their paper appeared in PLOS ONE in 2026. The model outlines organs so doctors can plan treatment. One model family reads fine detail but sees little of the whole scan. Another sees the whole scan but costs heavy computing power.

The new design runs both paths side by side. A fusion step weighs fine detail against broad shape at several scales. On two public scan sets, the model matched expert outlines closely. One set held many organs. The other held the aorta and its branches. Tools like this could cut the long hand-drawing work in hospitals.

Catalog record

CREATORBo Li, Wei Zhou, Haijun Li
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE61223
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0345549

Browse the full library · About the Tower of Babel