1.        Background

The operating room (OR) is a high-risk clinical environment in which staff must act, interact, and communicate efficiently to carry out complex tasks safely. The future OR can be conceptualised as a data ‘multiverse’ created from several ‘parallel universes’; these consist of the Patient, Surgeon, Staff and Devices/Environment.

Today, these are considered in isolation, when in reality they are interconnected and interdependent. Future surgical learning systems that optimise surgical performance across healthcare networks and along the entire patient journey, must therefore adopt a holistic approach that continually analyses data across the surgical data multiverse.

We postulate that events/ripples in one universe can be detected in constituent parallel universes, while causalities of events in the OR multiverse could be traced to individual parallel universes. Adopting this approach essentially requires unification of the human, physical, and digital entities in the OR under a single framework.

We propose to achieve this by integrating four central pillars:

  1. Holistic sensing of patient, surgeon, staff, and OR environment, with emphasis on human-centric sensing;
  2. AI based on multimodal data for enhanced cognition, and actions;
  3. Robotics leveraged by AI and perceptual user interfaces for augmenting task performance;
  4. Cloud architecture with big-data processing capacity, for integration of the above key elements and deployment at scale.


MAESTRO is the instrument that allows us to peer into and interact with the OR multiverse. It is driven by our vision for Transformative Healthcare, which aims to lay the foundations for the operating room of the mid-21st Century, a surgical environment powered by trustable, human-understanding artificial intelligence able to continually adapt and learn the best way to optimise safety, efficacy, teamwork, and ultimately clinical outcomes.

By means of its two modules, Observer and Director, MAESTRO supports the OR staff before, during and after a surgical procedure by:

  • Sensing the OR multiverse in its manifold aspects, including neglected phenomena such as staff physiological responses, visual behaviour and focus of attention, brain functions and cognitive workload, as well as adverse events that may escape the staff’s attention.
  • Assessing individual and team performance in real-time, longitudinally and across surgeries.
  • Orchestrating and assisting the surgical team via automated checkpoints, virtual/augmented visualisations, warnings, individualised and broadcasted alerts, automation, and other means.
  • Augmenting and optimising individual and collective operational capabilities, performance, and task ergonomics, through novel human-device interfacing modalities and robotics.



Multimodal sensor data-streams are collected over an interconnectivity backbone, anonymised and synchronised. Synchronised streams are transferred over a secure connection to the Cloud, where an AI continual learning process takes place for adaptation to new sites, partnering teams, ORs, and procedures. Multimodal situational awareness is achieved through advanced AI methodologies for enhanced perception and cognition and assigned to the Observer module. AI-driven guidance and assistance based on clinically relevant considerations, is delegated to the Director module. 

Our long-term aims are to demonstrate impact on real, co-designed clinical use cases and piloting at scale in partner hospitals, to explore technical, regulatory and ethical implications, and to boost trust and acceptability among all relevant stakeholders. Our ambition is to develop a MAESTRO ecosystem, both nationally and internationally, involving relevant stakeholders to support this ground-breaking vision.