AMBIOM

EfficientBioAI: New Open-Source Software Makes AI Models Lighter and Greener

Artificial intelligence has become an integral part of research. However, the better and more complex the models become, the higher their energy consumption. Researchers at ISAS and Peking University have therefore developed ready-to-use and open software that compresses existing bioimaging AI models. With the help of the new toolbox, scientists can now run their models faster and with significantly lower energy consumption.

Biomedical Image Analysis with AI

Dr Jianxu Chen is taking up a professorship in AI for Biomedical Analysis at the University of Duisburg-Essen. With his research group at ISAS, he is developing AI methods that can be used to comprehensively analyse enormous quantities of complex microscopy images.

It's What's Inside That Counts: Neutrophils Form Distinct Subtypes

Neutrophils can be divided into two distinct subgroups with completely different functions. The surface protein CD177 plays a decisive role in this. For the first time, researchers from the Faculty of Medicine at the University of Duisburg-Essen and ISAS have been able to demonstrate, using a new integrated proteomics and lipidomics analysis, that the molecular signatures in CD177-defined neutrophils differ significantly from one another. The presence of this surface protein has implications for disease progression, for example in head and neck tumours.

Four perspectives on a shared discovery

Although CD177⁺ and CD177⁻ neutrophils look identical under the microscope, they are in fact not the same. Four of the authors of “CD177 Deficiency Defines a Stable Subtype of Human Neutrophil Granulocytes with Tumour-Promoting Activity” (Advanced Science) explain how they identified the functional differences between these immune cells and which combined analytical methods made their discovery possible in the first place. This insight into their research demonstrates how interdisciplinary collaboration can reveal previously undiscovered information.

3 Questions for Lukas Fu

How can AI support the analysis of biomedical images? This was the question Lukas Fu explored during his school internship at ISAS. The 15-year-old also successfully completed his first programming project.

Separating Fluorescence Signals Faster and More Precisely with AI

When different fluorescence signals overlap during microscopy, their clear assignment to specific biological structures becomes difficult. To address this common problem, Dr Davide Panzeri is developing AI-based signal separation methods. The biophysicist has been awarded a Marie Skłodowska-Curie Fellowship by the European Union for his promising research project.

ISAS BFF Uploader

New Framework for Efficient Image Data Analysis in Biomedicine

How can AI be used optimally in bioimaging research? A team of international researchers and scientists led by Dr Jianxu Chen from ISAS has developed a framework designed to make it easier for biomedical scientists to work with data-centric AI. Using the example of vascular segmentation, the authors of the publication in the journal npj imaging now demonstrate the advantages of the new framework.

Humboldt Fellow Prof Dr Xiaowei Xu started at ISAS in March

Prof Dr Xiaowei Xu from the Chinese Guangdong Cardiovascular Institute is researching the clinical application of artificial intelligence (AI) in the context of cardiovascular diseases. For 18 months in total, he will be researching various AI methods for analysing cell images as a Humboldt Fellow at ISAS.

Valuable Connections: Dr Jianxu Chen

Dr Jianxu Chen leads the junior research group AMBIOM - Analysis of Microscopic BIOMedical Images at ISAS since 2021. He previously worked at the Allen Institute for Cell Science in Seattle, USA. In the ISAS -Kompakt series ‘Valuable Connections’, the computer scientist reports on his move to health research and to Germany.

What are you doing at ISAS, Leon?

What do marshmallows and chocolate have to do with cell analysis? Leon is finding out the answer during his school internship at ISAS. For ISAS Kompakt, the 15-year-old talks about what else he is learning during his time at the institute.

Valuable Connections: Adrian Sebuliba

Adrian Sebuliba joined the ISAS junior research group AMBIOM in 2023 as a software engineer. Previously, he worked for a digital commerce platform for the chemical industry in Uganda. In the ISAS Kompakt series ‘Valuable Connections’, he reports on his move into health research, among other things.