Our AI ensures the best distinction between tumor and stroma, so that tissue structures are easy to detect and diagnostic results are comprehensible
Our AI works on a single-cell basis mimicking the procedure of a pathologist when examining histological sections. Pathologists can comprehend and track which cell was classified how by the AI
We have developed proprietary clinical-grade deep learning architectures that are tailored to pathology and achieve results on par with experts. Combining human experts with our AI leads to accuracies beyond current standards
Image analysis tools need to earn the trust of the experts. Modern methods in explainable AI (xAI) help to open the black box and are an essential component to achieve this trust. We use such methods to provide the needed guidance to pathologists and clinicians
In addition to labeled data, there is a wealth of images without annotations in pathology. We harvest this treasure and extract relevant patterns even from unannotated data by using specialised methods such as self-supervised learning and generative adversarial networks
We are a lead author of the German "Guideline for the development of deep learning based image analysis systems in medicine" by the German Institute for Standardization (DIN) "DIN SPEC 13266 & 13288: Leitfaden für die Entwicklung von Deep-Learning-Bilderkennungssystemen in der Medizin"). We have advised several members of the German parliament.
We solve these challenges by building systems based on artificial intelligence (AI) that work out of the box. Our 0-click solutions are the first ever commercially available products that are able to instantly visualize the detected tumor cells in arbitrary images of IHC-stained breast cancer tissue without the need for manual fine-tuning.