Haematopathology
Free download
No. 3 of 20
Getting Started with AI in Hematopathology
A practical primer for the practising pathologist — what the technology actually does in haematopathology, what to ask of it, and how to start safely.
Hematopathology is the most data-integrative discipline in the laboratory. A single diagnosis of acute myeloid leukaemia routinely fuses morphology, immunophenotype by flow cytometry, cytogenetics, and a next-generation sequencing panel — then places the result against a classification system that is itself being rewritten around genomics. The American Society of Hematology’s 2025 position paper in Blood framed this precisely: the diagnosis and treatment of haematologic disorders depend on integrating imaging, path
What the primer covers
- SectionWhy hematopathology is a different problem
- SectionThe task map for haematology
- SectionWhere to begin
- SectionYour first 90 days
- SectionValidation specifics
- SectionFailure modes to anticipate
- SectionWhat to watch next
- SectionKey references
Who it is for
- Pathologists in haematopathology meeting diagnostic AI for the first time
- Laboratory leads assessing vendor claims and planning validation
- Trainees who want the regulatory and methodological context, not marketing
Educational guidance, not clinical protocol or regulatory advice. Algorithm authorisations and validation requirements change — confirm the current position with your regulator and your own quality system.