Ecker 2022 Abstract Bioblast
|5.5. «10+5» https://doi.org/10.26124/bec:2022-0001|
Ecker Rupert (2022)
Event: Bioblast 2022
In September 2021 the United States’ Food and Drug Administration (US-FDA) has approved the first artificial intelligence- (AI)-based decision support system for prostate cancer diagnostics. This hallmark indicates a historic decision as it is the first time in the history of medicine that a regulatory body has accepted a software-only solution, which analyses microscopic images by using artificial intelligence! This indicates both, that technologies reach a performance and maturity level that makes diagnostic routine applications not only possible but also feasible and that the market demand for such solutions has reached a level where it has become viable for industry to invest in the development of commercial solutions as a return on investment can be expected.
Our research teams at TissueGnostics and Queensland University of Technology have joined forces to combine TissueGnostics’ existing tissue cytometry technology platform and established knowhow with innovative AI solutions to establish The Virtual Histopathologist.
Tissue Cytometry permits to determine the in situ phenotype of cells as well as histological entities, like glands, vessels or tumor foci. Applications include but are not limited to the exploration of the cellular/tumor microenvironment and/or the spatial organization of cellular subpopulations, assessment of different bone structures, quantification of blood vessels and neovascularization as well as analysis of samples in multiplexing or multispectral mode.
Earlier attempts to analyse single cells in tissue have mostly been subject to visual estimation, or – at best – to manual counting for decades. Hence, experts usually had the choice of the “least of evils” between guessing and endless (manual) counting. In (tumor) immunology, infiltrating inflammatory cells need to be phenotypically characterized on a quantitative basis. To better understand the function of inflammatory cells in tumor development, type and number of inflammatory cells and their proximity to glandular/tumor structures have to be analyzed in situ and correlated with disease state. Using TissueFAXS™ Cytometry the time-consuming and error-prone human evaluation of stained histological sections can be approached with an observer-independent and reproducible technology platform, offering a high degree of automation, paired with user interaction at relevant points of the analytical workflow. This platform can be applied as a means of tissue cytometry for both immunofluorescence and immunohistochemistry and thus constitutes the microscopic equivalent to flow cytometry (FACS).
The TissueFAXS Cytometry platform incorporates Machine & Deep Learning algorithms and can be used in clinical multi-center studies to determine the immune response to certain drugs in situ, measure proliferation, apoptosis, cytokine expression, signalling molecules, and others. It can do end-point assays as well as live-cell imaging and time-kinetic experiments. TissueFAXS Cytometry also promotes tissue cytometry to a new level of quality, where complex cellular interactions can be addressed on the single-cell level but still in histological context.
• Bioblast editor: Plangger M
- TissueGnostics GmbH, Vienna, Austria
- School of Biomedical Sciences, Queensland University of Technology, Brisbane, Australia
List of abbreviations, terms and definitions - MitoPedia
Labels: MiParea: Instruments;methods Pathology: Cancer