Safe Haven Artificial Intelligence Platform (SHAIP)

  • Safe Haven Artificial Intelligence Platform
  • Technology
  • The Safe-Haven Artificial Intelligence Platform (SHAIP) team are a growing team of software engineers and architects who collaborate with the universities of Glasgow and Aberdeen, the NHS, Queen Elizabeth University Hospital and Aberdeen Royal Infirmary.

    They are responsible for building a set of services and web applications that include useful tools for clinicians to select and annotate patient data for machine learning, together with infrastructure for data scientists to develop, train and validate algorithms within the hospital environment. Once an algorithm has been created using SHAIP, it can be deployed into a Clinical Cockpit to allow effortless demonstration to clinicians.

    With some oversight from Japan, this team mainly collaborates within the UK and is a sub-team of the                        'Industrial Centre for Artificial Intelligence Research for Digital Diagnositcs (iCAIRD)'., a pan-Scotland collaboration of 15 partners from across academia, the NHS and industry.

    Here’s what our employees have to say:

    “I enjoy working with such a wide range of technologies on a variety of projects and I’m always learning. I’m excited that I have the opportunity to develop and learn from a great team and feel proud that we’re building infrastructure that helps support the NHS.” Daniel, Software Team Lead

  • See below the latest innovation from the Safe Haven AI Platform (SHAIP) team

    Ground Truth Application

    The Ground Truth application is a component of the SHAIP system used for annotation of medical images. Clinicians and researchers annotate regions of the dataset using the tool and then the annotations can then be used by data scientists for the development of machine learning algorithms. This tool was used by the Canon Medical Research Europe Image Analysis team to collect ground truth data for development of an algorithm to detect ischemic stroke and perform ASPECTS scoring.