Open access publication: An AI-enhanced care pathway planning and scheduling system

It is clear that Artificial intelligence (AI) is transforming the optimization of clinical and patient workflows in healthcare. But there is a need for research to specify clinical requirements for those systems to improve human-AI interaction in machine learning applications. Our partners from OYS, worked hard jointly with the five clinical partners during the first year of the project to produce the Functional Requirements specifications for our Smart Hospital Care Pathway Engine. These aimed to describe the core consortium clinical partners’ practical challenges and recognize opportunities, which form the basis of the project’s AI solution development.

Based on that work, the aim of this study was to assess content validity and prioritize the most relevant functionalities of an AI-enhanced care pathway planning and scheduling system, providing a comprehensive list of functionalities that can be implemented to design future AI-enhanced solutions, as well as evaluate the designed solutions against requirements.

They conducted a prospective content validity assessment using an electronic survey. The content of the survey was based on the clinical requirements but formulated into generic statements of required AI functionalities. Weighted ranking points served to prioritize the most relevant functionalities.

The statements concerning the AI functionalities were considered relevant, which might be due to the low level of organizational readiness for AI in healthcare.

The highest-ranked functionalities for an AI-enhanced care pathway planning and scheduling system were related to risk assessment, patient profiling, and resources, while the highest-ranked functionalities for the user interface were related to the explainability of machine learning models.

Published in BMC Health Services Research.
Read the publication here: http://bit.ly/3C7yedX