Page 28 - Read Online

Page 191 Body et al. Art Int Surg 2022;2:186-94 https://dx.doi.org/10.20517/ais.2022.28

self-criticism can be interpreted quickly. This may improve the time taken to acquire expert procedural
skills. It will also facilitate the attainment of educational tools for future surgeons.

Another application of ML in video analytics is surgical instrument recognition. The instrument used
[61]
during an operation provides a guide for which procedural step or action is underway . Algorithms
created by analysing laparoscopic gastrectomy found that 14 different surgical instruments could be
[62]
classified with an accuracy of 83.75% . ML applications were similarly able to recognise the surgical tool
used during MIS colorectal resections . Instrument recognition allows for instrument tracking, leading to
[62]
automated gesture and error identification. This application is particularly important as a potential
assessment of surgical expertise and skill . When using ML algorithms to analyse and track instrument
[63]
motion from surgical videos, surgical expertise was predicted with an accuracy of 83%, using the Objective
Structured Assessment of Technical Skills and Global Evaluative Assessment of Robotic Skills scores as the
[64]
standard . Deep learning neural networks trained on videos of MIS suturing could also classify a surgeon’s
[64]
suturing proficiency with an accuracy of over 80% . These models consitute the starting point for the
autonomous assessment of surgical competence. Similarly, computer vision models trained on over 2000
live robotic suturing videos could accurately identify the presence and type of suturing gesture (area under
the curve 0.88 and 0.87, respectively) . This may be the start of autonomous robotic suturing, an area of
[65]
surgery that may well transform the traditional concept of operating and surgical training.

Another novel and exciting application of ML in surgery is anatomical landmark recognition. A deep
learning model was trained to identify safe and unsafe zones of dissection in laparoscopic cholecystectomies
with an accuracy of 0.94 and 0.95 . The models could also produce a dynamic overlay of the zones onto
[66]
surgical videos. Similarly, deep artificial neural networks were created that can assess the achievement of the
[67]
critical view of safety during cholecystectomy before ligation . Taking this one step further, the Institute of
Image-guided Surgery of Strasbourg, France, has used real-time intra-operative augmented reality feedback
during cholecystectomy, highlighting the cystic duct and cystic artery. Expanding on these algorithms, there
is the potential to provide intra-operative augmented reality feedback to surgeons, particularly trainees,
thereby enabling decision support and enhancing patient safety. This future application may well result in a
real-time dynamic overlay of anatomy during the majority of robotic procedures.

Discrepancies in the technical abilities of surgeons are well established, and we know that higher proficiency
scores, assessed using video analysis, reduce patient morbidity . Robotic surgery allows for an increasing
[68]
number of objective performance metrics [69,70] , which, combined with ML algorithms, could be used as
automated surgical assessment tools in the future.

SUMMARY
Robotic surgery is rapidly expanding across surgical specialties. While it remains a low-volume technique in
pancreatic surgery, involvement in focused training programmes is recommended to enable surgeons to
master skills prior to independent operating. Training should be standardised to ensure the attainment of
assessment benchmarks and should include virtual simulation basic training in addition to procedural-
specific training. Exposure to basic robotic training should be implemented in the early postgraduate years.
Procedural techniques should be standardised to improve patient safety, theatre efficiency and the
continuation of robotic practice. Centres undertaking MIS pancreatic resections should consider the
implementation of dedicated team training and perform at least 20 robotic pancreatic resections per year.
Auditing outcomes for quality assurance and participation in MIS HPB international registries are
advocated. With the advance of robotic surgery and availability of surgical video datasets, ML in surgery
may rapidly increase. Combined ML applications and real-time dynamic augmented reality overlays will