Multimodal learning for scalable representation of high-dimensional medical data
Areej Alsaafin, Abubakr Shafique, Saghir Alfasly, Krishna R. Kalari, H. R. Tizhoosh · 2026
Read Multimodal learning for scalable representation of high-dimensional medical data by Areej Alsaafin, Abubakr Shafique, Saghir Alfasly, Krishna R. Kalari, H. R. Tizhoosh (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/pmc-12891988.txt.
Description
Areej Alsaafin, Abubakr Shafique, Saghir Alfasly, Krishna R. Kalari, and H R Tizhoosh present a study published in 2025 on fusing medical data types into one searchable code. Pathology slides and genomic profiles each hold part of a patient's story, but their formats clash, and most models read only one at a time.
Their framework, MarbliX, learns to compress both into short binary codes in a shared space, trained so matching patients land near each other. Similar cases can then be retrieved instantly, supporting case-based reasoning. In lung cancer the fused codes scored eighty-five to eighty-nine percent across metrics, beating either source alone. In kidney cancer the full-value codes scored best, with binary codes close behind.
Catalog record
| CREATOR | Areej Alsaafin, Abubakr Shafique, Saghir Alfasly, Krishna R. Kalari, H. R. Tizhoosh |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |