Maria Grammatikopoulou, Evangello Flouty, Abdolrahim Kadkhodamohammadi, Gwenol'e Quellec, Andre Chow, Jean Nehme, Imanol Luengo, Danail Stoyanov
Video feedback provides a wealth of information about surgical procedures and is the main sensory cue for surgeons. Scene understanding is crucial to computer assisted interventions (CAI) and to post-operative analysis of the surgical procedure. A fundamental building block of such capabilities is the identification and localization of surgical instruments and anatomical structures through semantic segmentation. Deep learning has advanced semantic segmentation techniques in the recent years but is inherently reliant on the availability of labelled datasets for model training. This paper introduces a dataset for semantic segmentation of cataract surgery videos complementing the publicly available CATARACTS challenge dataset. In addition, we benchmark the performance of several state-of-the-art deep learning models for semantic segmentation on the presented dataset. The dataset is publicly available at https://cataracts-semantic-segmentation2020.grand-challenge.org/.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (test) | 66.76 | UPN |
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (val) | 74.2 | UPN |
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (test) | 66.64 | HRNetv2 |
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (val) | 72.4 | HRNetv2 |
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (test) | 63.23 | DeepLabv3+ |
| 2D Semantic Segmentation task 3 (25 classes) | CaDIS | Mean IoU (val) | 68.6 | DeepLabv3+ |