At the recent European Hematology Association (EHA) Congress, the buzz word was ‘Artificial Intelligence’ (AI). Held in Milan, Italy from June 12-15, 2025, the Congress was a melting pot of scientific exploration and technological innovation, with AI technologies taking center stage. This is a testament to the growing impact of AI on healthcare and, more specifically, hematology. The spotlight was firmly on how the application of AI can revolutionize the clinical landscape and improve patient outcomes.
In one of the most impressive displays of AI’s potential, researchers from Juntendo University in Japan showcased a novel AI-driven approach known as ‘ghost cytometry.’ This ingenious blend of AI and flow cytometry is seen as a game-changer in the early diagnosis of chronic myeloid leukemia (CML) and the prediction of treatment response. The researchers’ work garnered attention due to its high F1 score of 0.79 in distinguishing between peripheral leukocytes of CML patients and healthy individuals. This indicates a high level of accuracy in the AI model, which could revolutionize CML diagnosis and treatment.
This breakthrough is poised to reshape the treatment landscape of CML. By predicting treatment response at an early stage, clinicians can tailor the treatment plan to individual patients, increasing the chances of a successful outcome. Furthermore, it guides the selection of tyrosine kinase inhibitors, a class of drugs used in CML treatment, further personalizing patient care.
But AI’s potential in hematology extends beyond CML. The Congress also highlighted other areas in hematology where AI is making waves. For instance, AI is being deployed in image analysis in bone marrow diagnostics and predicting biomarkers. This has already started to yield high-quality evidence, notably in the setting of nonhematological cancers.
However, this AI revolution is not without its challenges. Jakob Nicholas Kather, MD, MSc, of the Technical University of Dresden and University Hospital of Dresden, warned of the need for clinician literacy in AI technology. As healthcare is the most common justification for large financial investments in AI, it falls upon medical professionals to ensure that society reaps the benefits of these investments.
Although there are regulator-approved AI-enabled devices, their uptake has been slow. Kather noted the greater “shadow use” of nonapproved applications such as ChatGPT, which are useful for generating basic summaries or extracting value from unstructured real-world data but are not suitable for handling sensitive patient data.
The road to integrating AI fully into hematology will undoubtedly have its hurdles. However, the potential benefits, as demonstrated by the Juntendo University researchers and other studies presented at the EHA Congress, are too significant to ignore. The promise of early diagnosis, personalized treatment plans, and improved patient outcomes make a compelling case for embracing AI in hematology.
The journey of AI in healthcare has been a rollercoaster ride, with moments of great excitement and periods of disappointment. But the recent strides in hematology suggest we are on the cusp of a new era, where AI is not just hype but a powerful tool that can reshape the medical landscape. As the AI wave sweeps over the healthcare industry, the EHA Congress has shown that hematology is ready to ride the tide.
Read more from ajmc.com
