Surgical Workflow Recognition: From Analysis of Challenges to Architectural Study

Abstract

Algorithmic surgical workflow recognition is an ongoing research field and can be divided into laparoscopic (Internal) and operating room (External) analysis. So far, many different works for the internal analysis have been proposed with the combination of a frame-level and an additional temporal model to address the temporal ambiguities between different workflow phases. For the External recognition task, Clip-level methods are in the focus of researchers targeting the local ambiguities present in the operating room (OR) scene. In this work, we evaluate the performance of different combinations of common architectures for the task of surgical workflow recognition to provide a fair and comprehensive comparison of the methods for both settings, Internal and External. We show that the methods particularly designed for one setting can be transferred to the other mode and discuss the architecture effectiveness considering the main challenges for both Internal and External surgical workflow recognition.