Generating crossmodal gene expression from cancer histopathology improves multimodal AI predictions
Samiran Dey, Christopher R. S. Banerji, Partha Basuchowdhuri, Sanjoy K. Saha, Deepak Parashar, Tapabrata Chakraborti · 2025
Read Generating crossmodal gene expression from cancer histopathology improves multimodal AI predictions by Samiran Dey, Christopher R. S. Banerji, Partha Basuchowdhuri, Sanjoy K. Saha, Deepak Parashar, Tapabrata Chakraborti (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/pmc-12783123.txt.
Description
Samiran Dey and five co-authors show how AI can fill a missing data gap in cancer care in this two thousand twenty-five paper. Combining tissue images with gene activity data sharpens cancer grading and risk forecasts. But hospitals rarely order gene tests, so the second data type is usually absent. The team instead generates gene expression from the tissue image itself. Their model learns the link from cases where both exist.
They tested the idea on four cancer cohorts from two public collections. Adding the generated gene data clearly improved grading and risk estimates. Scores with faked data came close to those with real gene data. The trick could bring multimodal AI into clinics that only have slides. Pathologists and oncologists stand to gain.
Catalog record
| CREATOR | Samiran Dey, Christopher R. S. Banerji, Partha Basuchowdhuri, Sanjoy K. Saha, Deepak Parashar, Tapabrata Chakraborti |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |