To overcome this, researchers developed the largest open-access database on microscopic images of bone marrow cells to date.
“On top of our database, we have developed a neural network that outperforms previous machine learning algorithms for cell classification in terms of accuracy, but also terms of generalizability,” says Christian Matek, lead author of the new study.
The deep neural network is a machine learning concept specifically designed to process images. The analysis of bone marrow cells has not yet been performed with such advanced neural networks. This is because high-quality, public datasets have not been available until now.
Researchers are now aiming to expand their bone marrow cell database further to capture a broader range of findings and to prospectively validate their model.
The database and the model are freely available for research and training purposes – to educate professionals or as a reference for further AI-based approaches e.g. in blood cancer diagnostics.