The increasing administrative requirements place a considerable burden on medical professionals. The preparation of medical reports is particularly time-consuming, as specialist staff often transcribe voice recordings manually or sometimes still write reports by hand. Artificial intelligence (AI) offers great potential: modern speech recognition systems and large language models (LLMs) can efficiently transcribe, structure and improve the quality of reports. However, the use of AI-supported solutions raises key data protection issues - particularly with regard to professional secrecy and the use of cloud services. It is also difficult for many AI providers to assess when such systems are considered medical devices and what regulatory requirements are associated with them. As part of the Innovation Sandbox for AI, the Office for Economy of the Canton of Zurich and the Center for Information Technology, Society, and Law (ITSL) of the University of Zurich have developed recommendations for the safe use of AI technologies in medical documentation together with a large number of experts. The results should help to reduce the administrative workload in the healthcare sector and at the same time take into account the highest data protection and security standards. The guide is aimed in particular at providers of AI solutions, but can also provide important information for hospitals, medical practices and other healthcare providers.

Case studies from the innovation sandbox for artificial intelligence (AI)
The company MPAssist served as a case study within the innovation sandbox for AI. The organization submitted a project proposal to the sandbox in summer 2024. MPAssist offers AI solutions for medical reporting. The contents of this report were developed based on this specific case study.

With professional support from

Dr. André Baumgart
Head of Quality Management and Patient Safety, VZK

Dr. Nadine Bienefeld
Privatdozentin, ETH Zurich

Michèle Hess
Lawyer for digitization projects, Health Department of the Canton of Zurich

Dr. Rolf Kaufmann
Senior Medical Device Expert

Raffaele Lugli
Head of Process Digitization & Innovation, Canton of Zurich Health Department

Dr. med. Michael Neugebauer
Senior Physician & IT Officer, University Children's Hospital Zurich

Corinne Spirig
Chief Operating Officer, digital health center bülach (dhc)

Sebastian Svetel
Chief Information Security Officer, University Children's Hospital Zurich

Dr. med. Dr. phil. nat. Atanas Todorov
Chief Medical Officer, Arcondis

Peter Waldner
Head of eHealth, Department of Health, Canton of Zurich

Innovation sandbox for AI
The project team developed this document as part of the Innovation Sandbox for AI. The sandbox is a test environment for the implementation of AI projects from various sectors. The broad-based initiative from administration, business and research promotes responsible innovation by the project team and participating organizations working closely on regulatory issues and enabling the use of novel data sources. The contents of this report are not legally binding and do not represent an official position of public bodies. Any liability for legal aspects is excluded.

Report

German

English