RAIQC, which is leading the nine-partner consortium, has announced success in winning an award from UK Research & Innovation (UKRI) to help accelerate both the deployment and evaluation of medical imaging AI tools to healthcare providers including the NHS.
Over recent years, several artificial intelligence (AI) systems have been developed to aid with medical image interpretation. However, despite a number of systems being commercially available, their uptake and adoption into clinical practice has been slow. One reason for this is that development and validation of AI algorithms is a time-consuming process. Furthermore, pre-purchase evaluation and subsequent on-going quality assurance of these algorithms within the NHS is difficult to achieve.
RAIQC Ltd has already developed a web-based clinical simulation platform with combines DICOM viewing capability with structured reporting and annotation tools. The platform has been used to perform validation studies on several AI algorithms, both for commercial and academic partners. Through this project, RAIQC aims to further develop this platform into a solution that can address all the above market needs and will both accelerate the adoption of AI algorithms into clinical practice and allow ongoing performance assessment.
NDORMS has joined the consortium as an academic partner with RAIQC. RAIQC offer its platform's functionality to test AI algorithms, which have been developed during its DPhil studies.
Abhinav Singh and Rachel Kuo, NIHR Doctoral Research Fellows at NDORMS, said: 'The collaboration with RAIQC has been valuable to understand how clinicians perceive and interact with AI and how this influences decision-making. We have used the platform to evaluate diagnostic accuracy and confidence when using our assistive algorithms for detecting developmental dysplasia of the hip (DDH), and in assessing the use of radiological reports for identifying wrist or scaphoid fractures. The algorithms have been developed in close collaboration with the Department of Computer Science at the University of Oxford.'