This review summarises potential applications of artificial intelligence (AI) in histopathological diagnostics of leukemias reported in the available literature. It compares existing AI models focused on analysing image-based cellular morphology, trained on datasets of either blood smears or bone marrow slides. Key findings indicate a rising trend in research output and literature on that topic, with models achieving high accuracy rates — up to over 95% in leukemia detection and subtype classification. The implications suggest that AI can significantly enhance diagnostic precision, reduce subjectivity, and streamline workflows in hematopathology. Possible limitations and difficulties of introducing AI to routine diagnostics are also elaborated on. Overall, integrating AI into leukemia diagnostics holds promise for improving early detection, supporting clinical decision-making, and advancing treatment in haematological malignancies.
Czapliński et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: