Deep learning predicts IDH1 mutation status in acute myeloid leukemia from bone marrow smears
Acute myeloid leukemia (AML) with IDH1 mutations is amenable to targeted therapy, highlighting the need for reliable identification. We developed an attention-based multiple-instance learning model to predict IDH1 mutation status using bone marrow smear whole-slide images from 436 AML patients across four centers in Germany and the US, achieving a mean single pooled ROC-AUC of 0.81. In top tile analysis, model predictions for IDH1 were associated with immature leukemia phenotypes.
Authors
- Karsten Spiekermann (ORCID: https://orcid.org/0000-0002-5139-4957)
- Ishan Srivastava (ORCID: https://orcid.org/0000-0003-4113-9693)
- Tim Schmittmann
- Christoph Röllig (ORCID: https://orcid.org/0000-0002-3791-0548)
- Martin Bornhäuser (ORCID: https://orcid.org/0000-0002-5916-3029)
- Klaus Hans Metzeler (ORCID: https://orcid.org/0000-0003-3920-7490)
- Jan‐Niklas Eckardt (ORCID: https://orcid.org/0000-0002-3649-2823)
- Jan Moritz Middeke (ORCID: https://orcid.org/0000-0003-3250-293X)
- Torsten Haferlach (ORCID: https://orcid.org/0000-0003-0196-2837)
- Sebastian Riechert
- Christian Pohlkamp (ORCID: https://orcid.org/0000-0002-0653-3356)
- Susann Winter (ORCID: https://orcid.org/0000-0002-4228-4537)
- Katja Sockel (ORCID: https://orcid.org/0000-0003-1732-7299)
- Tobias Herold (ORCID: https://orcid.org/0000-0002-9615-9432)
- Ann‐Kathrin Eisfeld (ORCID: https://orcid.org/0000-0001-6442-1419)
- Andrew Srisuwananukorn (ORCID: https://orcid.org/0000-0002-8736-8726)
- Martin M. K. Schneider (ORCID: https://orcid.org/0000-0002-5501-4050)
- Markus Badstübner
- Chethan Babu V. Reddy
- Frank Kroschinsky
- Miriam Stoerk
- Maryam Schübel
Institutions
- German Cancer Research Center (DE)
- The Ohio State University Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (US)
- Helmholtz-Zentrum Dresden-Rossendorf (DE)
- LMU Klinikum (DE)
- München Klinik Bogenhausen (DE)
- Munich Leukemia Laboratory (Germany) (DE)
- München Klinik (DE)
- University Hospital Leipzig (DE)
- University Hospital Carl Gustav Carus (DE)
- Nationales Centrum für Tumorerkrankungen Dresden (DE)
- Else Kröner Fresenius Center for Digital Health (DE)
- German Cancer Consortium (DKTK) Dresden (DE)
- Technische Universität Dresden (DE)
- Ludwig-Maximilians-Universität München (DE)
Publication Details
- Journal
- npj Precision Oncology
- Published
- 2026-10-29
- DOI
- https://doi.org/10.1038/s41698-026-01762-x
- Primary Topic
- Digital Imaging for Blood Diseases
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Servier Deutschland