Using deep learning to better understand blood disorders
October 2019
Helmholtz Zentrum München, München, Germany(1)
LMU Munich, Munich, Germany(2)
LMU Munich, Munich, Germany(2)
The authors created a data set containing 18,000 images of individual leukocytes taken from 100 patients diagnosed with acute myeloid leukaemia and from 100 control patients. The specimens were digitized and used to train a deep learning convolutional neural network for leukocyte classification. The network classifies the most important cell types with high accuracy and identifies pathologies with human-level performance. This approach holds the potential to be used as a classification aid for examining much larger numbers of cells in a smear than can usually be done by a human expert. This will allow clinicians to recognize malignant cell populations with lower prevalence at an earlier stage of the disease.
Human-level recognition of blast cells in acute myeloid leukaemia with convolutional neural networks
Carsten Marr(1), Karsten Spiekermann(2)
Christian Matek et al. Nature Machine Intelligence 2019 [1]
Cancer Imaging Archive [2]
Chemie.de [3]
Added on: 12-17-2020
[1] https://www.nature.com/articles/s42256-019-0101-9[2] https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=61080958[3] https://q-more.chemie.de/q-more-artikel/329/mit-deep-learning-blutkrankheiten-besser-verstehen.html
