Real-Time Image-based Smoke Detection in Endoscopic Videos (bibtex)
@InProceedings{Leibetseder2017b, title = {Real-Time Image-based Smoke Detection in Endoscopic Videos}, author = {Leibetseder, Andreas and Primus, Manfred Jürgen and Petscharnig, Stefan and Schoeffmann, Klaus}, booktitle = {Proceedings of the on Thematic Workshops of ACM Multimedia 2017}, year = {2017}, address = {New York, NY, USA}, editor = {Wu, Wanmin and Yag, Jiancho and Tian, Qi and Zimmermann, Roger}, month = {jan}, pages = {296--304}, publisher = {ACM}, series = {Thematic Workshops '17}, abstract = {The nature of endoscopy as a type of minimally invasive surgery (MIS) requires surgeons to perform complex operations by merely inspecting a live camera feed. Inherently, a successful intervention depends upon ensuring proper working conditions, such as skillful camera handling, adequate lighting and removal of confounding factors, such as fluids or smoke. The latter is an undesirable byproduct of cauterizing tissue and not only constitutes a health hazard for the medical staff as well as the treated patients, it can also considerably obstruct the operating physician's field of view. Therefore, as a standard procedure the gaseous matter is evacuated by using specialized smoke suction systems that typically are activated manually whenever considered appropriate. We argue that image-based smoke detection can be employed to undertake such a decision, while as well being a useful indicator for relevant scenes in post-procedure analyses. This work represents a continued effort to previously conducted studies utilizing pre-trained convolutional neural networks (CNNs) and threshold-based saturation analysis. Specifically, we explore further methodologies for comparison and provide as well as evaluate a public dataset comprising over 100K smoke/non-smoke images extracted from the Cholec80 dataset, which is composed of 80 different cholecystectomy procedures. Having applied deep learning to merely 20K images of a custom dataset, we achieve Receiver Operating Characteristic (ROC) curves enclosing areas of over 0.98 for custom datasets and over 0.77 for the public dataset. Surprisingly, a fixed threshold for saturation-based histogram analysis still yields areas of over 0.78 and 0.75.}, doi = {10.1145/3126686.3126690}, isbn10 = {978-1-4503-5416-5}, keywords = {cnn classification, deep learning, endoscopic surgery, image processing, smoke detection}, language = {EN}, location = {Mountain View, California, USA}, talkdate = {2017.10.27}, talktype = {registered}, url = {http://doi.acm.org/10.1145/3126686.3126690} }
Powered by bibtexbrowser (with ITEC extensions)