SENCA-st: Integrating Spatial Transcriptomics and Histopathology with Cross Attention Shared Encoder for Region Identification in Cancer Pathology
Shanaka Liyanaarachchi, Chathurya Wijethunga, Shihab Aaqil Ahamed, Akthas Absar, Ranga Rodrigo · 2025
Read SENCA-st: Integrating Spatial Transcriptomics and Histopathology with Cross Attention Shared Encoder for Region Identification in Cancer Pathology by Shanaka Liyanaarachchi, Chathurya Wijethunga, Shihab Aaqil Ahamed, Akthas Absar, Ranga Rodrigo (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-2511-08573.txt.
Description
Shanaka Liyanaarachchi and four co-authors build a model that reads two kinds of cancer tissue data at once: gene activity maps from spatial transcriptomics and structure from pathology images. The goal is to mark tissue regions accurately, including tumor substructures tied to drug resistance.
Earlier methods leaned too far toward one data type, either drowning in transcriptomic noise or smoothing away functional detail. Their architecture, SENCA-st, uses a shared encoder with neighborhood cross-attention so both data types keep their own features. It highlights spots that look alike in images but differ in gene activity. The authors report it beats prior methods at finding tumor variation and micro-environment regions.
Catalog record
| CREATOR | Shanaka Liyanaarachchi, Chathurya Wijethunga, Shihab Aaqil Ahamed, Akthas Absar, Ranga Rodrigo |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-08573