05 February 2024

A method based on artificial intelligence (AI) can be used to predict genetic features of acute myeloid leukaemia (AML), according to a German study.

A team of IT specialists and physicians at the University of Münster and the University Hospital Münster used AI analysis on high-resolution microscopic images of bone marrow smears. This could enable clinicians to make decisions about precise treatment on the day of the diagnosis, without the need to wait for genetic analyses, they say.

The findings are published in Blood Advances.

They extracted genetic aberrations directly from extremely high-resolution multi-gigabyte scans from whole bone marrow smears taken from more than 400 AML patients.

The scans had resolution of 270,000 × 135,000 pixels on average, and one image was several gigabytes in size. This enabled them to extract more than two million single-cell images, they report.

Professor Benjamin Risse, who headed the work on the IT algorithmic developments, said: “We developed a new type of deep learning method, fully automatic, which was trained for a complex task by means of machine learning technology.

“In our case, the basic algorithm can automatically recognise the genetic features and the very fine patterns in big cytological images. The method then filtered the single-cell images into categories of different cell types, and it also showed any genetic aberrations.

“Interestingly, several patterns recognised by the algorithm cannot be identified by human observers. This is for example because the patterns may be too faint or because extremely fine textures are involved which remain hidden to the human eye, despite excellent imaging.”

One key advantage is in the end-to-end AI pipeline, which enables the interim results to be monitored and reduces to a minimum the manual preliminary work often necessary for machine learning, they report.

This is made possible by combining unsupervised, self-supervised and supervised learning processes. The first two processes require no manual data selection but try to extract relevant content automatically from the image data instead.

The team say the new method cannot replace genetic analyses, but it helps at a very early stage in the diagnostic clarification process for a leukaemia patient, by providing an idea of which genetic aberrations might underlie the disease.


Kockwelp J, Thiele S, Bartsch J, Haalck L, Gromoll J, Schlatt S, Exeler R, Bleckmann A, Lenz G, Wolf S, Steffen B, Berdel WE, Schliemann C, Risse B, Angenendt L. (2023) “Deep learning predicts therapy-relevant genetics in acute myeloid leukemia from Pappenheim-stained bone marrow smears.” Blood Advances, doi: 10.1182/bloodadvances.2023011076.

Link: https://ashpublications.org/bloodadvances/article/8/1/70/498751/Deep-learning-predicts-therapy-relevant-genetics


Disclaimer: The news stories shared on this site are used as a way to inform our members and followers of updates and relevant information happening in Haematology. The BSH does not endorse the content of news items from external sources, and is not in a position to verify the findings, accuracy or the source of any studies mentioned. Any medical or drugs information is provided as an information resource only, and is not to be relied on for any diagnostic or treatment purposes.

News service provided by Englemed News.