Authors:
Timo Brune
1
;
Björn Brune
2
;
Sascha Eschborn
2
and
Klaus Brinker
1
Affiliations:
1
University of Applied Sciences Hamm Lippstadt, Germany
;
2
Olympus Surgical Technologies Europe, Germany
Keyword(s):
Computer Vision, Feature Detection, Surf, Sift, Registration, Machine Learning, Supervised Learning, Endoscopes, Disinfection.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Cardiovascular Technologies
;
Computing and Telecommunications in Cardiology
;
Data Engineering
;
Decision Support Systems
;
Decision Support Systems, Remote Data Analysis
;
Development of Assistive Technology
;
Health Engineering and Technology Applications
;
Health Information Systems
;
Knowledge-Based Systems
;
Pattern Recognition and Machine Learning
;
Software Systems in Medicine
;
Symbolic Systems
Abstract:
This paper presents a complete analyzing system for detecting incorrect endoscope adaptions prior to the
use of chemical disinfection devices to guarantee hygienic standards and to save resources. The adaptions
are detected visually with the help of an image registration algorithm based on feature detection algorithms.
On top of the processing pipeline, we implemented a k-nearest neighbor algorithm to predict the status of
the adaption. The proposed approach shows good results in detecting the adaptions correctly.