Platform supporting an integrated analysis of image and multiOMICs data based on liquid biopsies for tumor diagnostics
VISIOMICS aims at developing a comprehensive solution for refined tumor diagnostics offering workflow management, multi-level data integration and advanced user interface solution in one efficient and certifiable workflow. Neuroblastoma, as a rare disease, will serve as a model to help solving the currently frequently observed problems in diagnostics accompanying personalized treatment strategies, i.e. integration of multi-level data despite of scarcity of samples and incomplete datasets. We therefore plan to expand our analysis by integrating DNA/RNA-sequencing and SNP-array data and combine them with imaging derived morphological and antibody staining properties of disseminated tumor cells isolated from liquid biopsies. Integration is performed in a sequential manner to better define genes and/or cell features, allowing to distinguish between samples from relapse versus non-relapse patients already at the time of diagnosis.
The VISIOMICS software platform, linked to a centralized database storing raw and processed data, will be an important tool enabling efficient user interaction and data visualisation tasks to translate expert knowledge into clinical/diagnostic analysis workflows.
Aim of the Project
Consortium
Publications
- Hamid Eghbal-Zadeh, Lukas Fischer, Niko Popitsch, Florian Kromp, Sabine Taschner-Mandl, Teresa Gerber, Eva Bozsaky, Peter F. Ambros, Inge M. Ambros, Gerhard Widmer and Bernhard A. Moser. An End-to-End Deep Neural Network with Attention-Based Localization for Breakpoint Detection in Single-Nucleotide Polymorphism Array Genomic Data. Journal of Computational Biology, vol. 26 (6), pp. 1-25, 2019.
- Shiva Alemzadeh, Florian Kromp, Bernhard Preim, Sabine Taschner-Mandl, Katja Bühler. A Visual Analytics Approach for Patient Stratification and Biomarker Discovery. Proc. of Eurographics Workshop on Visual Computing for Biology and Medicine (EG VCBM), pp. 91-96, 2019.
- Taha A. A., Bampoulidis A., Lupu M. Chance influence in datasets with a large number of features. Data Science – Analytics and Applications, P21—26. Proceedings of the 2nd International Data Science Conference – iDSC2019.
- Florian Kromp, Lukas Fischer, Eva Bozsaky, Inge Ambros, Wolfgang Doerr, Sabine Taschner-Mandl, Peter Ambros, Allen Hanbury. Deep Learning Architectures for Generalized Immunofluorescence Based Nuclear Image Segmentation. ArXiv preprints, 2019.
- F. Kromp et al., Evaluation of Deep Learning architectures for complex immunofluorescence nuclear image segmentation. in IEEE Transactions on Medical Imaging, 2021.
- Filip Mivalt. Segmentation of Phase Contrast Images in Multi-Epitope Ligand Cartography for Image Quantification at the Single Cell Level. Masterthesis, FH Technikum Wien, 2019.
- Daria Lazic. Deep Multi-Epitope Imaging of the Bone Marrow Disseminated Disease in Neuroblastoma. Masterthesis, TU Wien, 2019.
- Florian Kromp. Machine learning for tissue image analysis. Dissertation, TU Wien, 2019.
-
Florian Kromp, Eva Bozsaky, Fikret Rifatbegovic, Lukas Fischer, Magdalena Ambros, Maria Berneder, Tamara Weiss, et al. An Annotated Fluorescence Image Dataset for Training Nuclear Segmentation Methods. Scientific Data 7, no. 1 (December 2020): 262.
Posters
Deep learning based tool to analyze I-FISH spots in consecutive sections of heterogeneously amplified neuroblastoma tumors
ANR2018 – Advances in Neuroblastoma Research 2018, San Francisco – https://www.anrmeeting.org/meetings-2018.phpTowards characterizing new biomarkers for disseminated tumor cells and the microenvironment of the metastatic bone marrow niche in stage m neuroblastoma
SIOPEN 2018, Israel – https://www.siopenisrael.com/VISIOMICS – Platform supporting an integrated analysis of image and multiOMICs data for biology based advanced tumor diagnostics
From Lab to Life Symposium 2018, Vienna – https://www.labtolife.at/Characterizing new tumor biomarkers and the microenvironment of the metastatic bone marrow niche in stage M neuroblastoma using quantitative imaging and deep-learning based feature extraction
From Lab to Life Symposium 2018, Vienna – https://www.labtolife.at/
Acknowledgements
The project VISIOMICS was funded by the Austrian Research Promotion Agency (FFG) via the COIN program under the grantno. 861750.