iCAIRD is promoting 2 distinct eco-systems for AI development. One focusses on the needs of AI in radiology and the other on AI in pathology.
SHAIP & Clinical Cockpits
Operating within the NHS national Safe-haven environment, Canon’s Safe-Haven AI Platform (SHAIP) and Clinical Cockpit systems will enable secure AI development without export of identifiable data. Continuous integration between SHAIP and the Clinical Cockpit will allow AI algorithms to be deployed in real-time providing rapid feedback and informing clinical decisions. The aim is to streamline the setup of training environments for machine learning and running experiments to enable clinicians & researchers to quickly determine their success. This will reduce barriers to SMEs, simplify job scheduling & enable establishment of a network of regional centres for AI development and validation.
Philips equipment is targeting digital capture of pathology imaging to a HDRUK national pathology image archive which is open-standard, scalable and accessible to SMEs. Philips will provide dedicated AI tools and pipelines, working in concert with the Edinburgh Parallel Computing Centre (EPCC) to complete digital conversion of the Glasgow Pathology Department and incorporate established clinical workflows to accelerate development of deep learning. Images from multiple sources will be stored in both proprietary, open iSyntax and TIF formats to provide flexibility in data access and algorithm development.
OMERO is an enterprise data management platform which enables access, processing, sharing, analysis & (where appropriate) publication of scientific image data & metadata.
NVIDIA® DGX-1™is the integrated software and hardware system that supports commitment to AI research with an optimized combination of computer power, software and deep learning performance. Nvidia is assisting the SHAIP and Clinical Cockpit environment establishment with provision of their DGX-1™ demonstrator, DGX-1™ technology support and deep learning architecture solution building.