LTM-UNet: Linear Transformer–Mamba with Attention-Based U-Net for Context-Aware Breast Ultrasound Image Segmentation
Shivpratap Singh Kushwah, Santosh Prakash Chouhan, Narinder Singh Punn, Mahua Bhattacharya · 2026
Read LTM-UNet: Linear Transformer–Mamba with Attention-Based U-Net for Context-Aware Breast Ultrasound Image Segmentation by Shivpratap Singh Kushwah, Santosh Prakash Chouhan, Narinder Singh Punn, Mahua Bhattacharya (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/PMC13298203.txt.
Description
Outlining tumors in breast ultrasound scans demands both the big picture and fine edges, and Shivpratap Singh Kushwah with co-authors built a network for exactly that in 2026. Their model, LTM-UNet, blends a transformer encoder with a state-space decoder inside the familiar U-shaped segmentation design.
Attention-guided skip connections carry fine spatial detail across the network while limiting the mismatch between encoder and decoder features. On benchmark sets the model scored Dice coefficients of 82.41% on BUSI and 86.62% on Dataset B, beating several existing approaches on overlap measures as well. The authors credit the mix of global context and direction-aware decoding for the gains.
Catalog record
| CREATOR | Shivpratap Singh Kushwah, Santosh Prakash Chouhan, Narinder Singh Punn, Mahua Bhattacharya |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65679 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |