A secure platform for analysis of AI reporting tools
Researchers, academics, and AI developers can make use of the RAIQC platform to run image-based studies or generate clinical evidence for UK, FDA and EMA regulatory approval.
A one-stop shop for your next reader study
RAIQC can support you with your next project from conception to completion. We have extensive experience in:
-
Study design, analysis and write-up
-
Data sourcing, annotation and expert ground truthing
-
Sourcing radiologists and other clinicians for Multi-Reader Multi-Case studies
Case Studies
Lunit CXR
RAIQC supports evaluation of Lunit INSIGHT CXR AI tool for enhancing clinician chest X-ray interpretation
GE Healthcare PTX AI
RAIQC used for the evaluation of GE Healthcare's AI tool for the detection of pneumothoraces
Take part in reader studies or research projects
All studies require participants, so if you’re interested in taking part in forthcoming projects you can let us know by updating your RAIQC profile to indicate this, as well as listing any specialisms to help match readers to studies. If or when a suitable opportunity arises we’ll get in touch.
Running a reader study for an AI Tool
A study or research project will be tailored to your exact needs, but an example work flow might have three stages.
-
The initial read
Reporters review a selection of cases with any findings recorded by the RAIQC platform. Confidence in any diagnosis can also be recorded, together with how long reporting each case has taken. -
The second read
Following a ‘wash out’ period, reporters look at the cases again, this time with a secondary capture, generated by the AI Tool made available alongside the primary one. -
Statistical evaluation
With both reads completed, comparisons can be made and analysis can be carried out to reveal any performance benefits to using AI: Are reporters more accurate? Quicker? More confident?