Methods and datasets for segmentation of minimally invasive surgical instruments in endoscopic images and videos: A review of the state of the art

Tobias Rueckert, Daniel Rueckert, Christoph Palm

Abstract: Highlights•Identification of commonly used datasets and quantification of their frequency of use.•Systematic search on three platforms for surgical instrument segmentation methods.•Analysis of semantic and instance-based methods with and without temporal information.•Discussion of the reviewed literature and identification of existing shortcomings.•Highlighting of available potential for future developments.