“Artificial intelligence” generated leukemia blast cells through pre-annotated digital images by using the Motic Infinity 60 at the Tan Tock Seng Hospital, Singapore.
This has enabled the creation of large image databanks of blood cells, allowing the exponential growth of artificial intelligence (AI) studies in blood cell classification.
Image 1. (A) Blood cells in myeloid leukemia stained with Ehrlich's triple stain, as seen under a microscope. Watercolor after A. Goodall, 1912. Wellcome Collection. https://wellcomecollection.org/works/dtfntjgs/images?id=h5k9mkax. (B and C) Leukemic blasts in the background of red cells. (Wright stain, 80× objective, Motic EasyScan Infinity 60). (D–F) Artificial Intelligence generated leukemic blasts and red blood cells using the StyleSwin generative adversarial network (GAN) algorithm. https://github.com/microsoft/StyleSwin.
A total of 21.739 pre-annotated digital images derived from peripheral blood films of patients with acute myeloid leukemia (Wright stain, 80x objective, MoticEasyScan Infinity 60) were used in the training and evaluation process.
AI holds great promise in medicine due to its ability to effortlessly assimilate, integrate, and generate vast amounts of clinical data. However, AI needs suitable digital scanners for the digitalization of samples such as the Motic Infinity 60 of the MoticEasyScan series to facilitate the training tasks that will make its enhancement possible.
