What question did this study set out to answer?

The research aimed to redefine myeloid neoplasms classification by integrating genomic data with clinical information.

December 8, 2025Blood

Redefining the myeloid disease continuum: Machine learning–based development and validation of an integrated multimodal classification of myeloid neoplasms

View Full Paper

Ask AI

Bookmark

Authors

Machine learning reveals distinct features of myeloid neoplasms, suggesting improved genomic classification and clinical outcomes.

The research aimed to redefine myeloid neoplasms classification by integrating genomic data with clinical information.
Developed a multimodal classification model using machine learning techniques on a discovery cohort of 6484 patients.
Validated the model with an international cohort of 2012 patients.
Utilized unsupervised random forest algorithms and PCA for data integration and dimensionality reduction.
Identified distinct clusters and subclusters among myeloid neoplasms, highlighting their heterogeneous nature.
Predicted risks for clonal progression and survival based on integrated features.
Created a pan-MN prognostic score that outperformed existing tools with a c-index of 0.84.

Pérez‐Encinas et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark