We are proud to announce that Vironova has been awarded funding for a two-year research project entitled “Interactive visualization tools for verification and improvement of deep neural network predictions.” The project will stretch over two years (Jan 2018- Dec 2019) and represents yet another collaboration of strategical importance to Vironova in the areas of Artificial Intelligence (AI) and Machine Learning, applied to image analysis in future healthcare.
The project is part of the Analytic Imaging Diagnostic Arena (AIDA), an arena for research and innovation on image-based diagnostics and part of the national Strategic Innovation Program, Medtech4Health, a joint initiative by VINNOVA, Formas and the Swedish Energy Agency. Vironova’s Ida-Maria Sintorn is project leader and the project is run together with collaborating partners, Kjell Hultenby (Electron Microscopy Unit, Huddinge University hospital) and pathologist Anca Dragomir (Uppsala University Hospital).
Understanding and being able to show what leads to a decision using AI and machine learning methods is important in gaining acceptance for their use in clinical diagnostics. The project will adapt and develop interactive visualization tools to examine the importance of regions and details of an image in decision making by an artificial neural network. The tools developed in this project will not be directly used clinically but rather support and serve as a catalyst for deploying other decision support systems in the clinic.
For Vironova this project adds invaluable depth and breadth to our already front-line expertise in applying AI to image analysis. Experience from our research collaborations is transferred continuously into the analysis methods behind our MiniTEM and VAS platforms, allowing us to maintain an unmatched leadership in our field.
You can find more information on AIDA here:
Other related news: Vironova proud sponsor of first Swedish symposium on Deep Learning
Vironova presents on artificial intelligence data driven revolution
Vironova’s ultrastructural pathology collaboration
Läs mer på MyNewsDesk