A survey of the machine learning techniques and their role in intelligent and autonomous surgical actions

Advances in technology and computing play an increasingly important role in the evolution of modern surgical techniques and paradigms. In this article we review the current role of machine learning (ML) techniques in the context of surgery with a focus on surgical robotics and we provide a perspective on the future possibilities for enhancing the effectiveness of procedures by integrating ML in the operating room. We focus our review on ML techniques directly applied to surgery, surgical robotics, and surgical training and assessment. The widespread use of ML methods in diagnosis and medical image computing is beyond the scope of the review. We performed searches on PubMed and IEEE Explore using combinations of the keywords: ML, surgery, robotics, surgical and medical robotics, skill learning, skill analysis, learning to perceive. Studies making use of ML methods in the context of surgery are increasingly reported. In particular, there Y. Kassahun · B. Yu · A. T. Tibebu · J. H. Metzen Faculty 3 Mathematics and Computer Science University of Bremen, Robert-Hooke-Str. 5, D-28359 Bremen, Germany E-mail: bingbin.yu@uni-bremen.de D. Stoyanov Centre for Medical Image Computing and Department of Computer Science University College London, UK S. Giannarou Hamlyn Centre for Robotic Surgery Imperial College London, London, UK E. Vander Poorten Department of Mechanical Engineering University of Leuven Celestijnenlaan 300B, B-3001 Heverlee, Belgium is focus on using ML for developing the tools to understand and model surgical skill, competence, and surgical workflow. Some initial works have begun integrating the increased understanding of the surgical process into the control of recent surgical robots and devices. ML is an expanding field. It is widely used to efficiently process vast amounts of data and to interpret it for real-time decision making. Already widely used in imaging and diagnosis, we believe ML methods will also play an important role in surgery and interventional treatments. In particular ML could become a game changer into the conception of cognitive surgical robots. ML would allow extracting surgical skill, learned through demonstration by human experts, and transfer this to robotic skills as such offering intelligent surgical assistance. Such systems would significantly surpass the state of the art in surgical robotic technology, which, at present, merely plays the role of an instrument that enhances the surgeon’s dexterity.