Development of medical applications based on AI models and register data – regulatory considerations

Abstract

Artificial intelligence based methods, especially machine learning (ML), are increasingly used in healthcare for automatic medical image analysis and clinical decision support systems. Development and validation of ML models involve processing of large volumes of personal data. We analysed regulatory impacts on ML based application development especially from the perspective of privacy protection and usage of ML models as a basis for software under medical device regulation (MDR). We present best practices for ML application development and personal data usage in a use case of predicting elderly individuals’ future need for healthcare and social welfare services.