Explainable AI (XAI)

Developing methods that generate an XAI model and interface

Creating an Explainable Artificial Intelligence Framework to Increase Nurses’ Confidence in an Interhospital Transfer Scenario. In proceedings of the AAAI 2022 Workshop: Trustworthy AI for Healthcare, Virtual, March 2022.


Emergency patients’ Interhospital transfers (IHT) is becoming important due to adverse clinical outcomes and overcrowding of the transferred emergency department (ED). Since determining whether referred patients should be authorized needs deep consideration of multiple influencing factors, experienced ED patient flow nurses take the critical responsibility of making this decision in order to enhance the efficiency of ED management. Artificial Intelligence (AI) systems have been developed to support health providers’ decisions. However, it is questionable whether health providers are willing to follow the decisions recommended by such AI systems due to its unclear decision-making process. In this study, we propose an XAI framework to increase nurses’ confidence in the decision recommended by the AI system in an IHT scenario.

For more details, visit: pdf

An Explainable AI Tool to Support Decision-making in an Emergency Patients Transfer Request Scenario. In proceedings of the Korean Society of Medical Informatics 2021 (KOSMI’21), Yongin, Korea, November 2021.