Artefact detection in video endoscopy using retinanet and focal loss function
Oksuz I, Clough JR, King AP, Schnabel JA (2019)
Publication Type: Conference contribution
Publication year: 2019
Publisher: CEUR-WS
Book Volume: 2366
Conference Proceedings Title: CEUR Workshop Proceedings
Event location: Venice, ITA
Abstract
Endoscopic Artefact Detection (EAD) is a fundamental task for enabling the use of endoscopy images for diagnosis and treatment of diseases in multiple organs. Precise detection of specific artefacts such as pixel saturations, motion blur, specular reflections, bubbles and instruments is essential for high-quality frame restoration. This work describes our submission to the EAD 2019 challenge to detect bounding boxes for seven classes of artefacts in endoscopy videos. Our method is based on focal loss and Retina-net architecture with Resnet-152 backbone. We have generated a large derivative dataset by augmenting the original images with free-form deformations to prevent over-fitting. Our method reaches a mAP of 0.2719 and a IoU of 0.3456 for the detection task over all classes of artefact for 195 images. We report comparable performance for the generalization dataset reaching a mAP of 0.2974 and deviation from the detection dataset of 0.0859.
Involved external institutions
United Kingdom (GB)How to cite
APA:
Oksuz, I., Clough, J.R., King, A.P., & Schnabel, J.A. (2019). Artefact detection in video endoscopy using retinanet and focal loss function. In CEUR Workshop Proceedings. Venice, ITA: CEUR-WS.
MLA:
Oksuz, Ilkay, et al. "Artefact detection in video endoscopy using retinanet and focal loss function." Proceedings of the 2019 Challenge on Endoscopy Artefacts Detection: Multi-Class Artefact Detection in Video Endoscopy, EAD 2019, Venice, ITA CEUR-WS, 2019.
BibTeX: Download