Fakultät Informatik und Mathematik
Regensburg Center for Artificial Intelligence
Regensburg Center of Biomedical Engineering
Regensburg Center of Health Sciences and Technology

Prof. Dr. rer nat. Christoph Palm

DeepMIC - Deep learning based endoscope tracking

[1] https://endovissub2018-roboticscenesegmentation.grand-challenge.org/

Funding

Grant from funds of the Bavarian Research Foundation,
Funding code: AZ-1506-21

Period and volume

August 2021 to August 2024

Overall project: approx. 534 T€, sub-project of the OTH Regensburg: approx.  200 T€

Cooperation partners

Project description

The DeepMIC project is creating a new approach for an intelligent, collaborative assistance system for camera guidance during minimally invasive surgical procedures. The new assistance system is to be characterized by an adaptivity in use that has not yet been achieved to any degree, intuitive operability and the ability to cooperate actively (semi-) automatically with the surgeon and thus be capable of the best possible camera guidance virtually on its own.

The innovative approach consists of a continuous evaluation and classification of the information of the endoscopic camera image by methods of artificial intelligence (here especially Deep Learning) in combination with natural speech recognition. Combined with knowledge from the surgical workflow, the system should allow interaction with the surgeon that is similar to a human assistant and can thus react directly to the current requirements of the procedure.