[Back]

C. Gangl, M. Lackner, J. Maly, S. Woltran:
"Aggregating Expert Opinions in Support of Medical Diagnostic Decision-Making";
Talk: KR4HC 2019 - Knowledge Representation for Health Care/ProHealth, Poznan, Polen; 2019-06-26 - 2019-06-29; in: "Knowledge Representation for Health Care/ProHealth, KR4HC 2019", (2019), 56 - 62.

Medical doctors are often faced with challenging diagnostic decisions, which require the consideration of all eligible differential di­agnoses. Diagnostic decisions (i.e., which test to order next in a given situation) have a high impact, as non-targeted diagnostic strategies may cause delayed treatments. It is thus desirable for medical professionals to be able to tap into the knowledge from more experienced colleagues - a process which can be fostered and supported by knowledge-based soft­ware tools. In this position paper, we outline the potential and challenges of applying methods from Computational Social Choice (COMSOC) to aggregate expert advice on diagnostic strategies.
Gathering and aggregating expert opinions is a challenging task, espe­cially in the medical domain. We discuss the necessary requirements for COMSOC methods to be applicable in the diagnostic support setting, in particular requirements for opinion elicitation and opinion aggregation. The main goal of our research is to build a system that supports diag­nostic decision-making based on reliable expert knowledge. Principled methods and analyses from COMSOC guarantee that recommendations are reliable, sound, and explainable.

Medical Decision Support · Computational Social Choice · Preference Aggregation · Preference Elicitation · Multi-Winner Voting

Project Head Reinhard Pichler:
SuDeMa

Project Head Stefan Woltran:
START