Brian Booth, a postdoctoral researcher from the Image Processing and Interpretation Lab (UGent), joined industry leaders earlier this year to speak on the use of AI in additive manufacturing workflows. The first webinar, hosted by AI4Growth on March 25th, covered the use of AI regression models to detect specific manufacturing defects from high-speed video. The second webinar, hosted by Flanders Make and AI Flanders on April 29th, covered the use of autoencoders to identify anomalies in signal time series’.
The diagnosis of myelodysplastic syndromes (MDS) is challenging and mainly relies on bone marrow morphology and genetic tests. In collaboration with clinicians from VUMC Amsterdam, scientists from the Yvan Saeys lab (VIB-UGent Center for Inflammation Research) have developed a machine learning model to help in the automated diagnosis of MDS. This model resulted in a significantly better diagnostic power, while drastically reducing the time and resources needed for analysis.
Call for papers for several workshops in conjunction with the ECML-PKDD conference in September 2021
Several workshops will be organized in conjunction with the ECML-PKDD conference in September 2021.
You can now submit your papers.
Tony Belpaeme spoke at the European Commission’s Science for Policy Conference “What future for European robotics?” where he discussed how AI and new developments in robotics will shape the future of Human-Robot Interaction, together with Hae Won Park (MIT Media Lab) and Stephan Vincent-Lancrin (OECD).
A jury from the Research Foundation Flanders (FWO) and the Fonds de la Recherche Scientifique (FNRS) has recently selected the 2020 laureates of the AstraZeneca Foundation Awards.
Piet Demeester obtains Methusalem funding for long term research on next generation wireless networks (SHAPE)
Piet Demeester has received Methusalem funding for the project SHAPE: Next Generation Wireless Networks.
One of the first 15 iBOF projects granted to Tijl De Bie (UGent) and Luc De Raedt, Jesse Davis (KULeuven)
Within the first iBOF call the team of Tijl De Bie joined forces with Luc De Raedt and Jesse Davis from the KULeuven on the topic of “Automating Data Science: the Next Frontiers”.
Coinciding with the publication of the “Artificiële intelligentie voor lokale besturen” (artificial intelligence for local government) book by the Vlaamse Vereniging voor Steden en Gemeenten Tony Belpaeme (UGent-imec) gave a keynote at the online launch event, looking at what impact recent developments in AI have on addressing challenges faced by society.
Special Issue published by EOS magazine on Health and Technology highlights some research projects of AI.UGent
The special issue of the EOS magazine on Health and Technology focuses on a wide range of aspects of future healthcare.
The Chan Zuckerberg Initiative (CZI) announced $3.8 million in funding for 23 grants to support open-source software projects essential to biomedical research, enabling software maintenance, growth, development, and community engagement. Software by the lab of Prof. Yvan Saeys (VIB-UGent Center for Inflammation Research) for the analysis and visualization of single-cell data is one of the projects that will receive funding as part of CZI’s Essential Open Source Software for Science (EOSS) program.
What happens inside a cell when it is activated, changing, or responding to variations in its environment? Researchers from the VIB-UGent Center for Inflammation Research have developed a map of how to best model these cellular dynamics. Their work not only highlights the outstanding challenges of tracking cells throughout their growth and lifetime, but also pioneers new ways of evaluating computational biology methods that aim to do this.
Every year, Rxivist compiles a list of the most downloaded bioRxiv preprints. The organization has recently put together the list for 2018. And at number 10, we find a paper from the Yvan Saeys group at the VIB-UGent Center for Inflammation Research. This is the first time VIB research makes it into the top 10 of this list.
Tom Dhaene (IDLab) - Deep learning models are a common tool to identify complex patterns in data. However, they still requires tens of thousands of training samples. In this project we investigate the use of Bayesian models such as the (deep) Gaussian process for deep learning and generative modeling modeling when data is scarce.
Tom Dhaene (IDLab) - Generative models are one of the most promising approaches to learn the world around us. In this project we investigate their potential for the design of linear passive electronic systems. Generative models, an unsupervised machine learning technique, are applied to learn the complete design space and to generate interesting valid designs.