JMIR Publications

ABSTRACT

Background:

Artificial intelligence (AI)-based systems are receiving more and more attention in the healthcare sector. While the use of AI is well advanced in some medical applications, such as image recognition, it is still in its infancy in others, such as Clinical Decision Support Systems (CDSS). Examples for AI-based CDSS can be found in the context of sepsis prediction or antibiotic prescription. Scientific literature indicates that such systems can support physicians in their daily work and lead to improved patient outcomes. Nevertheless, there are various problems and barriers in this context which should be considered. Expert interviews with relevant stakeholders were conducted to explore potential problems and barriers as well as ways to improve the quality AI-based CDSS and their integration into care.

Objective:

We aimed to identify opportunities to optimize AI-based CDSS and their integration into care from the perspective of experts.

Methods:

Semi-structured online expert interviews were conducted. Experts representing the perspectives of patients, physicians, caregivers, developers, health insurance representatives, researchers (especially in the fields of law and IT), experts in regulation, market admission and quality management or assurance, and experts in the field of ethics were included. The conversations were recorded and transcribed. Subsequently, a qualitative content analysis was performed. The different approaches to improvement were categorized into groups ("Technology", "Data", "Users", "Studies", "Law", and "General"). These also served as deductive codes. Inductive codes were determined within an internal project workshop.

Results:

Thirteen individual and two double interviews were conducted with 17 experts. A total of 227 expert statements were related to ways to improve AI-based CDSS and their integration into care. Suggestions were heterogeneous and concerned various areas. Besides improvement of the systems, potential for optimization was seen on the user side as well as in the environment in which the systems are used. Many of the suggested improvements concerned different stakeholders from various domains.

Conclusions:

Multiple different approaches, to enhance AI-based CDSS and their integration, were identified. The suggestions concerned different points within the life cycle as well as various stakeholders. Therefore, interdisciplinary work can be seen as a key result. The effectiveness and accuracy of the potential match between problems in the context of AI-based CDSS and suggestions for care improvement should be investigated in further studies.