Deep learning differentiates myeloproliferative neoplasms and predicts JAK2 and CALR mutations from bone marrow smears
Jan-Niklas Eckardt, Chethan Babu V Reddy, Holger Hauspurg, Sebastian Riechert, Tim Schmittmann, Ishan Srivastava, Susann Winter, Katja Sockel, Markus Badstübner, Martin Bornhäuser, Markus Tiemann, Jan Moritz Middeke · 2026
Read Deep learning differentiates myeloproliferative neoplasms and predicts JAK2 and CALR mutations from bone marrow smears by Jan-Niklas Eckardt, Chethan Babu V Reddy, Holger Hauspurg, Sebastian Riechert, Tim Schmittmann, Ishan Srivastava, Susann Winter, Katja Sockel, Markus Badstübner, Martin Bornhäuser, Markus Tiemann, Jan Moritz Middeke (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-13518502.txt.
Description
Eckardt and co-authors teach a computer to tell apart three blood cancers from microscope images of bone marrow smears. The three diseases look alike under the microscope, and telling them apart guides treatment. Their 2026 model learned from 460 digitized smear slides, using features from a vision system that had trained itself on images without labels.
The model sorted the three diseases with solid accuracy. It could also guess, from the image alone, whether a tumor carried one of two key gene changes, with fair accuracy in two of the diseases. The authors say this hints that gene changes leave faint traces in cell shape. They suggest that mixing image data with lab and clinic data could sharpen the tool further.
Catalog record
| CREATOR | Jan-Niklas Eckardt, Chethan Babu V Reddy, Holger Hauspurg, Sebastian Riechert, Tim Schmittmann, Ishan Srivastava, Susann Winter, Katja Sockel, Markus Badstübner, Martin Bornhäuser, Markus Tiemann, Jan Moritz Middeke |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 12 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |