Recently published
AI prediction of acute leukemia subtypes from routine laboratory data
Clinical data science from routine care
A 2026 Nature Communications paper reports international testing and refinement of algorithms that predict acute leukemia subtypes from routine laboratory data. Related commentary appears in Lancet Digital Health (2024).
What problem are we trying to solve?
Can algorithms trained on routine laboratory parameters distinguish acute leukemia subtypes in a way that generalizes across hospitals and health systems?
Why it matters
Specialized leukemia diagnostics are unevenly available. A carefully validated model on widely available laboratory data could support earlier orientation of care, without replacing clinical diagnosis.
Approach
International testing and refinement of AI algorithms, as reported in Nature Communications 2026. Public copy must follow the paper, not press paraphrases.