Much of the research on eXplainable Artificial Intelligence (XAI) has centered on providing transparency of machine learning models. More recently, the focus on human-centered approaches to XAI has increased. Yet, there is a lack of practical methods and examples on the integration of human factors into the development processes of AI-generated explanations that humans prove to uptake for better performance. This paper presents a case study of an application of a human-centered design approach for AI-generated explanations. The approach consists of three components: Domain analysis to define the concept & context of explanations, Requirements elicitation & assessment to derive the use cases & explanation requirements, and the consequential Multi-modal interaction design & evaluation to create a library of design patterns for explanations. In a case study, we adopt the DoReMi-approach to design explanations for a Clinical Decision Support System (CDSS) for child health. In the requirements elicitation & assessment, a user study with experienced paediatricians uncovered what explanations the CDSS should provide. In the interaction design & evaluation, a second user study tested the consequential interaction design patterns. This case study provided a first set of user requirements and design patterns for an explainable decision support system in medical diagnosis, showing how to involve expert end users in the development process and how to develop, more or less, generic solutions for general design problems in XAI.

This study aimed to determine the most appropriate touch-based interaction technique for I2Vote, an image-based audience response system for radiology education in which users need to accurately mark a target on a medical image. Four plausible techniques were identified: land-on, take-off, zoom-pointing, and shift. The techniques were implemented in such a way that they could be used on any modern device. An empirical study was performed in which users marked a target on an image using all four techniques on either a smartphone or a tablet. The techniques were compared in terms of accuracy, efficiency, ease of use, intuitiveness, and compatibility with the different devices. The results showed that shift was the most accurate technique, but it was hampered by its high complexity and low intuitiveness. Land-on was the fastest technique but also the least accurate. Take-off and zoom-pointing provided the best trade-off between accuracy, efficiency, ease of use, and intuitiveness. We therefore conclude that both take-off and zoom-pointing are viable interaction techniques for I2Vote.