Real-time image processing and data recording

ENDOMIND is a computer-aided detection (CADe) prototype for real-time image capture and processing to highlight polyps and other findings during the streaming of images. It runs via a high-end computer that is able to capture different formatted video signals. The system can be connected using HDMI, DVI, SDI, and in analogue form to endoscope towers in order to transfer various signals from endoscopes to the CADe system. The system processes streams and outputs a new signal, more rapidly than human delay perception. This allows physicians to use it without any recognizable latency or disadvantages, while all the benefits of our AI systems such as polyp detection are available. This is possible through multiple parallel pipelines that distribute and simultaneously process the workload.

The image pipeline processes the most recently captured frames to produce a display as quickly as possible. Simultaneously, frames from previous cycles are processed through the AI pipeline, predicting boundary boxes, which are used as an overlay. Independently of time-critical processing, frames are both encoded and recorded, along with additional data, in a state-of-the-art fashion. The system is designed to anonymize videos by cropping and overlaying sensitive areas such as those containing patient names. The data collected open up new opportunities for improving on or building new neuronal networks and further enhancing the system.

Research for this project is part of the “Baden-Württemberg Health-Care Location Forum” (Forum Gesundheitsstandort Baden-Württemberg), sponsored by the Ministry of Science, Research and Art of the state of Baden-Württemberg.

The project has been developed in collaboration with Prof. Dr. Frank Puppe, Head of Chair for Artificial Intelligence and Knowledge Systems, Julius Maximilians University of Würzburg. Other participants include Prof. Dr. Dr. W.G. Zoller, Katharinenhospital Stuttgart, and Prof. Dr. J.F. Riemann, Stiftung LebensBlicke.

The software including the CADe system (ENDOMIND) and a detailed installation handbook is freely available for research purposes.

Publication

Lux TJ et al. (2022)
Pilot study of a new freely available computer‑aided polyp detection system in clinical practice.
International Journal of Colorectal Disease 2022.
Go to publication 
Software download link

For any questions and additional explanations please feel free to contact the author Alexander Hann. E-mail: hann_a@ ukw.de