Visual Assistance in Development and Validation of Bayesian Networks for Clinical Decision Support



Juliane Müller-Sielaff, Seyed Behnam Beladi, Stephanie W. Vrede, Monique Meuschke, Peter J.F. Lucas, Johanna M.A. Pijnenborg, Steffen Oeltze-Jafra

 View presentation:2022-10-21T14:12:00ZGMT-0600Change your timezone on the schedule page
2022-10-21T14:12:00Z

Exemplar figure, described by caption below — The common development process of Clinical Decision Support Models (CDSM) based on Bayesian networks (BN) requires medical researchers providing the domain expertise and a modeling expert developing the model. This collaborative but generally time-expensive procedure, however, hampers in model generation and updating. Furthermore, because the modelling expert might make design decisions on the experts’ domain, at some stage the medical expert may be confronted with a lack of understanding of the resulting BN model. To address these problems, we gathered the requirements on a visual approach allowing medical researchers to generate and validate CDSMs based on BNs mainly independently.

Prerecorded Talk

The live footage of the talk, including the Q&A, can be viewed on the session page, Visual Analytics of Health Data.

Fast forward

Keywords

Bayesian networks, Visual Analysis, Clinical Decision Support, Causal Model Development

Abstract

The development and validation of Clinical Decision Support Models (CDSM) based on Bayesian networks (BN) is commonly done in a collaborative work between medical researchers providing the domain expertise and computer scientists developing the decision support model. Although modern tools provide facilities for data-driven model generation, domain experts are required to validate the accuracy of the learned model and to provide expert knowledge for fine-tuning it while computer scientists are needed to integrate this knowledge in the learned model (hybrid modeling approach). This generally time-expensive procedure hampers CDSM generation and updating. To address this problem, we developed a novel interactive visual approach allowing medical researchers with less knowledge in CDSM to develop and validate BNs based on domain specific data mainly independently and thus, diminishing the need for an additional computer scientist. In this context, we abstracted and simplified the common workflow in BN development as well as adjusted the workflow to medical experts' needs. We demonstrate our visual approach with data of endometrial cancer patients and evaluated it with six medical researchers who are domain experts in the gynecological field.