Broadening the HTA of medical AI: A review of the literature to inform a tailored approach

Objectives As current health technology assessment (HTA) frameworks do not provide specific guidance on the assessment of medical artificial intelligence (AI), this study aimed to propose a conceptual framework for a broad HTA of medical AI. Methods A systematic literature review and a targeted search of policy documents was conducted to distill the relevant medical AI assessment elements. Three exemplary cases were selected to illustrate various elements: (1) An application supporting radiologists in stroke-care (2) A natural language processing application for clinical data abstraction (3) An ICU-discharge decision-making application. Results A total of 31 policy documents and 9 academic publications were selected, from which a list of 29 issues was distilled. The issues were grouped by four focus areas: (1) Technology & Performance, (2) Human & Organizational, (3) Legal & Ethical and (4) Transparency & Usability. Each assessment element was extensively discussed in the test, and the elements clinical effectiveness, clinical workflow, workforce, interoperability, fairness and explainability were further highlighted through the exemplary cases. Conclusion The current methodology of HTA requires extension to make it suitable for a broad evaluation of medical AI technologies. The 29-item assessment list that we propose needs a tailored approach for distinct types of medical AI, since the conceptualisation of the issues differs across applications.