RE-AIMED: Readjusted responses by use of AI in medical calls
Medical call centres are increasingly used to guide the population to the right level of health care. The operators at the call centres have a huge responsibility when choosing a response on behalf of the caller. The operators need correct and relevant information to make good judgements. Good communication with the caller is a pre-asset for access to such information. Today multitasking disturbs the operators` attentiveness to the callers, their workflow and their ability to communicate. Additionally, decision support tools, which were introduced to ensure the quality of the decisions made, tend to steer the conversation into predefined paths, and thereby reducing the quality of communication.
In RE-AIMED we will explore how the use of artificial intelligence (AI) can improve the communication between the operator and the caller, and the workflow of the operator. AI can use information from the conversation to provide the operator with questions adjusted to the natural flow of the conversation. The operator documents the call by choosing pairs of answers and questions, while simultaneously receiving help to recognize medical patterns. Relatively few calls are regarding medical emergencies, but these few calls must be discerned from less severe cases. A central challenge is therefore how to prevent the AI from misclassifying the few, severe cases. Another challenge is how to design a user interface that provides information and guidance to the operator without disrupting the operators` ability to focus and communicate with the caller.
The project will create large, standardized and detail-coded data sets, which describe medical calls. This allows for further research on telephone triage, medical decision-making, communication and reduction of biases in machine learning.