DAME (Deep learning Algorithms for Medical image Evaluation)

Entwicklung von Software-Algorithmen für die Detektion von Abweichungen in medizinischen Bildern (Imaging) auf der Grundlage von Machine Learning

The aim of this project is to explore software solutions using deep learning technology to achieve automatic, fast and reliable detection of abnormalities (such as cancer) in medical images. The main advantage of this innovation is the development of a generic algorithm to recognize patterns in images, independent of the type of image (CT, MR, etc) or type of abnormality. This allows to use the same software system to solve a multitude of different clinical problems. The goal is not only to quickly identify healthy individuals, but also to detect abnormalities that are not directly linked to the clinical question (incidental findings). By automatically identifying all abnormalities in the images, missing something crucial will be avoided.


Geplante Projektkosten

1.186.256,00 €


1.10.2017 - 1.10.2021


Erhöhung der grenzüberschreitenden Innovationskraft im Programmgebiet

Lead Partner

Universitair Medisch Centrum Groningen (UMCG)


COSMONIO IMAGING B.V., Use-Lab GmbH, Radiologie West-Münsterland , Universität Oldenburg (Pius-Hospital Oldenburg)


Finanzierer Betrag
Universität Oldenburg (Pius-Hospital Oldenburg) 3.636,00 €
Radiologie West-Münsterland 5.454,00 €
Provincie Groningen 50.890,00 €
Use-Lab GmbH 75.401,00 €
Ministerie van Economische Zaken en Klimaat 76.333,00 €
Universitair Medisch Centrum Groningen (UMCG) 126.729,00 €
MB Niedersachsen 127.223,00 €
COSMONIO IMAGING B.V. 127.462,00 €
EFRE / EFRO 593.128,00 €