Segmentation of Low-Grade Brain Tumors Using Mutual Attention Multimodal MRI
Hiroyuki Seshimo, Essam A. Rashed · 2024
Read Segmentation of Low-Grade Brain Tumors Using Mutual Attention Multimodal MRI by Hiroyuki Seshimo, Essam A. Rashed (2024) 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-11644188.txt.
Description
Seshimo and Rashed outline brain scans for slow-growing tumors with a network that lets two scan types teach each other where the tumor lies. These tumors spread diffusely with faint edges, so tracing them by hand is slow and error prone. Their 2024 model fuses two MRI sequences, T2-weighted and FLAIR, through mutual attention rather than simply stacking them.
Attention lets each sequence highlight what the other finds salient, sharpening borders where the tumor signal is subtle. On a public set of 35 tumor cases, the model reached an average overlap score of 0.87 against expert tracings. The two chosen sequences carried most of the value. The authors present the design as a step toward context-aware scan fusion in clinics.
Catalog record
| CREATOR | Hiroyuki Seshimo, Essam A. Rashed |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |