Next Generation Digital Morphology: Blast Preclassification in Bone Marrow Aspirates.
AI performs well at detecting cancer cells in blood smears, but bone marrow analysis remains too complex for clinical replacement of expert microscopy.
This narrative review evaluates AI-based digital morphology for blast detection in PB smears and bone marrow aspirates, finding that while deep learning achieves near-human performance for PB smear analysis (≥90% sensitivity/specificity), bone marrow aspirate analysis remains more variable due to cellular complexity. Current commercial platforms cannot replace manual microscopy for blasts, and ICSH recommends broad validation and regulatory approval before routine clinical integration.
What the study was
- Study design
- Narrative review
- Category
- Diagnostics
- Maturity
- Exploratory
- Journal
- International Journal of Laboratory Hematology
Why it surfaced
Useful synthesis of AI blast detection landscape for multiple watchlist topics (T1, T2, T4); no new primary data; ICSH position on regulatory prudence is notable.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.