Artificial intelligence in haematopathology: current perspective and future directions

Artificial intelligence (AI) is revolutionizing pathology by improving diagnostic accuracy and efficiency, particularly in pattern recognition within visual data. Whilst the role of AI in hematopathology is advancing, it is still behind the progress seen in solid tumor analysis. In areas such as lymph node and bone marrow pathology AI can support quantitative analysis, differential diagnosis, molecular prediction, and the exploration of novel biomarkers. However, challenges remain, including the need for extensive clinical validation, overcoming dataset limitations, and addressing potential overfitting issues. Future advancements will hinge on integrating both supervised and unsupervised learning techniques, improving digital pathology adoption to increase the availability of multicenter datasets, and harnessing emerging technologies such as spatial omics. Overall, AI is poised to augment pathologists' capabilities, offering more precise diagnostic tools and insights. However, the ultimate interpretation of data will remain a pathologist's domain, with AI-based methods serving as a complementary tool rather than a replacement for conventional microscopy.