Fig. 1: Overview of the current artificial intelligence (AI) strategy for classification and explanation of myeloproliferative neoplasm (MPN) disease prediction. | Leukemia

Fig. 1: Overview of the current artificial intelligence (AI) strategy for classification and explanation of myeloproliferative neoplasm (MPN) disease prediction.

From: Artificial intelligence differentiates prefibrotic primary myelofibrosis with thrombocytosis from essential thrombocythemia using digitized bone marrow biopsy images

Fig. 1

The AI framework was trained on digitized bone marrow biopsies from prefibrotic primary myelofibrosis (prePMF) and essential thrombocythemia (ET) patients treated at the University of Florence, Italy and validated on patients at Moffitt Cancer Center, Florida, United States. A generative AI model was developed to produce synthetic images of prePMF and ET disease states to depict highly relevant morphological features as determined by the classification algorithm. Finally, a survey of real and generated images was evaluated by human hematopathologists and compared with the AI classification model.

Back to article page