What is it about?

The KIPA project combines digital consent forms with an AI-based chat and voice system that patients can use before radiological procedures. It allows them to review information at home, ask questions, and receive explanations at their own pace. We interviewed six radiologists and six medical assistants from four German university hospitals and conducted usability testing. Participants were cautiously optimistic that KIPA could improve patient preparation, understanding, standardization, and workflow efficiency while reducing repetitive communication. However, they emphasized that informed consent remains a trust-based human process. Concerns included complex or emotional questions, digital literacy, data protection, response reliability, technical integration, and possible additional review work.

Featured Image

Why is it important?

Informed consent is ethically and legally essential, but discussions in radiology often take place under considerable time pressure. A carefully designed AI system could answer routine questions beforehand and enable clinicians to focus their limited consultation time on individual risks, concerns, and emotional support. Successful implementation requires more than an accurate chatbot: the system must integrate smoothly with clinical workflows, clearly communicate its limitations, protect patient data, accommodate different levels of digital literacy, and offer an easy transition to human assistance. Involving both patients and healthcare professionals in development is therefore crucial. AI should strengthen patient autonomy and professional communication rather than automate consent as a purely administrative task.

Perspectives

At the heart of informed consent is a relationship based on trust, understanding, and professional responsibility. For me, the value of an AI dialogue system lies not in imitating or replacing clinicians, but in giving patients more time to prepare and professionals more capacity for individualized communication.

Prof. Dr. Thomas Rigotti
Leibniz Institute for Resilience Research (LIR) and Johannes Gutenberg University Mainz

Read the Original

This page is a summary of: AI-Based Dialogue Systems for Informed Consent in Radiology: Professional Perspectives from the KIPA Study, May 2026, IOS Press,
DOI: 10.3233/shti260601.
You can read the full text:

Read

Contributors

The following have contributed to this page