VISIOMICS

Analysis platform for better prediction of relapse in childhood cancer

The probability of a relapse in individual cancers is estimated already at the time of diagnosis. The entirety of all genes of the cancer cell and their conversion into RNA is often analysed using bioinformatic methods, which leads to extremely large, so-called "OMICS" data sets. In addition, image analyses of microscopic images are carried out using methods of machine learning or artificial intelligence. The results are also included in the forecast.

Within the scope of this project, these approaches are to be combined using the example of neuroblastoma, a tumor of infancy and young children, in order to enable earlier and more precise detection of relapses.

To this end, more genomic material of the tumor from the blood or bone marrow ("liquid biopsies") is to be examined. The data obtained together with clinical and image data should lead to improved forecasting possibilities with the aid of machine learning methods (keyword artificial intelligence), better data visualization (visual analytics) and a platform for an interactive diagnostics workflow to be developed in the project.

SCCH is mainly involved in the VISIOMICS project in the field of image analysis. Here, SCCH researchers focus primarily on the segmentation and classification of cells using machine learning algorithms, in particular deep neuronal networks (deep learning, artificial intelligence) in close cooperation with colleagues at Labdia Labordiagnostik GmbH and the University Hospital Erlangen.

Funding

The project VISIOMICS is funded by the Austrian Research Promotion Agency (FFG) under the COIN programme line "Networks".

Contact

Lukas Fischer

Fischer Lukas

Research Manager Data Science
Phone: +43 50 343 828

back