Acute Myeloid Leukemia

Acute myeloid leukemia (AML) is a severe blood cancer requiring rapid and accurate diagnosis for effective treatment. Current research focuses on improving diagnostic methods using machine learning, particularly employing models like variational autoencoders and set-transformers to analyze flow cytometry and microscopic image data, aiming for automated and objective subtype classification. These advancements, including the use of optimal transport for dimensionality reduction and self-supervised learning for data-efficient model training, promise faster, more accurate diagnoses and improved risk stratification for personalized treatment strategies, ultimately impacting patient outcomes and clinical decision-making. Addressing biases in algorithms related to age and sex is also a critical area of ongoing investigation.