New evidence show that artificial intelligence (AI) – computer-based algorithms designed to recognize patterns automatically – could help identify two rare blood cancers known as prefibrotic primary myelofibrosis (prePMF) and essential thrombocythemia (ET). Distinguishing between these two blood cancers is difficult but critical to help guide treatment and enroll patients to clinical trials. Researchers with The Ohio State University Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC – James) report these findings today at the American Society of Hematology annual meeting in San Diego. Andrew Srisuwananukorn, MD, a physician in the division of hematology at OSUCCC – James, led the research while at the Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York. It is the first among several AI-focused hematology projects that Srisuwananukorn and his team will continue at Ohio State. “We hope that future versions of our algorithm can help accurately diagnose and guide treatment for patients with these rare blood cancers,” said Srisuwananukorn. “We imagine that this AI support tool could help pathologists and clinicians across cancer institutions, but particularly at centers that do not see these diagnoses frequently. Potentially, our algorithm may improve patient enrollment in clinical trials, leading to study results that better reflect the expected disease outcomes.” Understanding rare blood cancers PrePMF is rarer and has a much worse prognosis than ET, with a median survival of 12 years compared with 22 years for ET. As a result, prePMF may require more aggressive treatment, however, getting a definitive diagnosis from lab and biopsy interpretations is difficult. For this study, researchers used a model previously trained to recognize the general features of more than 32,000 cell samples. This is believed to be the largest test to use AI to differentiate between prePMF and ET. The AI-based system used patient images from the University of Florence, Italy (between 2007 and 2023) and the Moffitt Cancer Center, Tampa (between 2013 and 2022), and was able to return results in just over six seconds for a new patient, on average, with an overall accuracy of 92.3%. “We hope to relay to practicing clinicians that AI algorithms and models are rapidly being developed and can be quite helpful for our day-to-day practices,” said Srisuwananukorn. “We have developed the software to view features of a bone marrow biopsy to accurately define the subtle differences between prePMF and ET.” The hope is to carefully test the algorithm for prePMF and ET detection in larger groups of patients with similar disease at Ohio State and other medical centers to strengthen the accuracy for detection. Furthermore, these same AI frameworks can be developed to predict other outcomes of interest to physicians and patients, including risk stratification and response to therapy. To learn more about research and patient cancer care at the OSUCCC – James, visit cancer.osu.edu. To learn more about Ohio State’s presence at ASH, visit cancer.osu.edu/ASH2023. ### Media Contact: Mary Ellen Fiorino, mary.fiorino@osumc.edu