AI Evaluation

KiTEC:AIEU - The AI Centre Evaluation Unit

It conducts independent evaluations of AI-enabled medical devices, digital, and other health technologies for the NHS and NICE.

About Us

The AIEU, powered by KiTEC, assists start-ups and SMEs to develop and provide evidence and support for their value proposition.

The nature and complexity of AI-based technology brought about the establishment of a dedicated AI evaluation unit (AIEU) within the AI Centre. The AIEU is an integral part of KiTEC, a National Institute for Health and Care Excellence (NICE)-accredited Health Technology Assessment (HTA) centre with extensive experience evaluating innovative healthcare technologies.

Utilising the existing expertise in Kings's Health Partners (KHP) an Academic Health Sciences Centre, the AIEU provides a ‘one-stop shop’ for a systems-based approach to accelerate the bench to market development, validation, evaluation and adoption of AI-based technologies.

The AIEU has the infrastructure and expertise to assess AI technologies as per the NHS Digital Health Technology (DHTs) standards in:

High-performance GPU computing infrastructure
High-capacity imaging and non-imaging data storage
Ethical approval for a research database for the development and evaluation of AI technologies
FLIP - Federated Learning and Interoperability Platform
AIDE - AI Deployment Engine

Frameworks

The AIEU utilises national and international frameworks developed to guide the evidence generation strategies of complex interventions and digital health technologies such as the:

NICE's Evidence standard framework for DHTs
NHS Digital Health Technology (DHTs) standards
The European Medical Device Regulation and any developments arising from the newly establish Artificial Intelligence Medical Devices (AIMDs) branch of the International Medical Device Regulators Forum (IMDRF).

Current Evaluation Projects

KiTEC has recently completed the first independent evaluation of a medical device incorporating AI functions for the newly established DHTs programme.

As part of this evaluation, KiTEC collaborated with researchers from the AI Centre for a technical evaluation of ZIO XT, a patient-friendly heart monitor that uses deep learning to detect and classify arrhythmias with expert-level accuracy.

The evaluation investigated the algorithm its development, including training and adaptive updates, the algorithm outputs and report creation, and how it is integrated into the clinical pathways.

How to get involved

The AIEU is uniquely positioned to provide a wide variety of bespoke services relating to AI evaluations. If you are interested, then get in touch so we can discuss the specific needs of your project.