Home  | News

24.01.2025

Teaser image to How to analyze millions of individual cells

How to Analyze Millions of Individual Cells

TUM News

Researchers from the Technical University of Munich (TUM) and Helmholtz Munich have tested self-supervised learning as a promising approach for analysing over 20 million individual cells. MCML PI Fabian Theis holds the Chair of Mathematical Modelling of Biological Systems at TUM. Together with his team, he has investigated whether self-supervised learning is better suited to analysing large amounts of data than other methods.

Given the enormous amounts of data generated by advances in single-cell technology, it is important to interpret them efficiently in order to recognise differences between healthy cells and those with diseases such as lung cancer or COVID-19. Self-supervised learning does not require pre-classified data and enables the robust processing of large amounts of data.

The study shows that this method improves performance, especially on transfer tasks and zero-shot predictions. The researchers compare masked learning and contrastive learning and find that masked learning is better suited for large datasets. The results could lead to the development of virtual cells that map the diversity of cells in different datasets and are useful for analysing cell changes in diseases.

#research #theis
Subscribe to RSS News feed

Related

Link to Language Shapes Gender Bias in AI Images

30.10.2025

Language Shapes Gender Bias in AI Images

Alexander Fraser shows AI image generators reproduce gender stereotypes differently across languages, highlighting the need for fair multilingual AI.

Link to Björn Ommer appointed LMU Chief AI Officer

20.10.2025

Björn Ommer Appointed LMU Chief AI Officer

Our PI Björn Ommer has been appointed LMU’s Chief AI Officer to strengthen AI research and collaborations.

Link to

17.10.2025

MCML at ICCV 2025

MCML researchers are represented with 28 papers at ICCV 2025 (22 Main, and 6 Workshops).

Link to

17.10.2025

MCML at IROS 2025

MCML researchers are represented with 2 papers at IROS 2025.

Link to

16.10.2025

MCML at CSCW 2025

MCML researchers are represented with 1 paper at CSCW 2025.

Back to Top