Knowledge transfer for surgical activity prediction

TL;DR

This paper presents methods for improving surgical activity prediction by leveraging knowledge transfer techniques, including semantic encoding and transfer learning across datasets, resulting in significant accuracy improvements.

Contribution

It introduces a novel combination of semantic encoding and transfer learning approaches to enhance surgical activity prediction with limited data.

Findings

Transfer learning outperforms simple data combination.

22% improvement in activity prediction accuracy.

Semantic encoding accelerates learning process.

Abstract

Lack of training data hinders automatic recognition and prediction of surgical activities necessary for situation-aware operating rooms. We propose using knowledge transfer to compensate for data deficit and improve prediction. We used two approaches to extract and transfer surgical process knowledge. First, we encoded semantic information about surgical terms using word embedding which boosted learning process. Secondly, we passed knowledge between different clinical datasets of neurosurgical procedures using transfer learning. Transfer learning was shown to be more effective than a simple combination of data, especially for less similar procedures. The combination of two methods provided 22% improvement of activity prediction. We also made several pertinent observations about surgical practices.