Our research group comprises the following three areas contributing to novel concepts and testbed systems for visual adaptive consumer health information (A+CHIS).

Area 1: Knowledge Visualization for Adaptive Health Information

Description of Area 1 and its objectives

Knowledge Visualization for Adaptive Health Information covers concepts and implementations of adaptive visualizations of health data building on evidence-based health information and CHIS (area 2) and cognitive psychology principles in health information use (area 3).

Key goals of Area 1

A key goal is the development of approaches to determine appropriate interactive visual representations of health information automatically. This involves finding appropriate knowledge structures for the representation of information on health, the consumers, and visualization processes. Building on this, research is conducted into mechanisms that automatically decide what information, by which level of detail, and visual representation should be displayed based on the personal needs of the consumer.

Methods and approaches of Area 1

The approaches will build on known guidelines and user studies on the effectiveness of information visualization displays, and methods of machine learning and recommender systems to model and learn user interest. The latter will allow not only for consumer information according to the pull principle (consumer explicitly requests information), but also the push principle (further information is recommended). Testbed software will be implemented for evaluation and demonstration purposes. This area will be led by the Institute of Computer Graphics and Knowledge Visualization at Graz University of Technology.

Expertise and research experience of the leading institute

The institute has expertise in research of effective interactive data visualizations for different types of data and user tasks, including time-dependent measurement data, and network- and text-oriented data, all of which are relevant to A+CHIS. In addition, the institute has research experience in adapting visual data displays to user information needs, e.g., through the use of multimodal interactions such as eye tracking or relevance feedback, to infer and serve user interests.

Expected outcomes and contributions of Area 1

Our research contributes to understanding the adaptation of visual information for user needs, and contribute new adaptive visualization methods for medical information.

Area 2: Evidence-based Consumer Health Information

Description of Area 2 and its objectives

Evidence-based Consumer Health Information covers the study of existing CHIS, deriving principles and goals used in traditional (static, non-interactive) CHIS, on which we build on and improve on.

Key tasks of Area 2

These principles inform our new approaches, by incorporating known working methods, while extending and improving them for multidimensional adaptivity. Evidence-based medical information on type 2 diabetes mellitus is researched from the state of the art as the health information basis for our A+CHIS testbed implementation. Based on existing quality criteria for CHIS, we will also define standards for the methodological quality of the A+CHIS to ensure its trustworthiness. An additional task is the provision of testbed environments (e.g., general practice network) for evaluation, and the involvement of stakeholders from the medical domain in the feedback process, like general practice associations or health insurers.

Importance of evidence-based research for CHIS

A strictly evidence-based approach is required to ensure the presented information is of high quality. This is clearly necessary in view of the heterogeneity and differing levels of evidence presented in the medical literature, let alone the questionable quality of information available in the wilds of the Internet.

Role of the leading institute in Area 2

This research area is led by the Institute of General Practice and Evidence-based Health Services Research at the Medical University of Graz. The institute conducts evidence-based research into selecting and curating medical information that is suitable for CHIS.

Expected outcomes and contributions of Area 2

In the development of evidence-based health information systems and quality assurance for health information systems, we will build on the results of previous projects. Our research will help to increase the reliability and quality of health information and improve the adaptation of health information to suit its users.

Area 3: Cognitive Psychology of Adaptive Health Information Systems

Description of Area 3 and its objectives

Cognitive Psychology of Adaptive Health Information Systems covers cognitive aspects involved when consumers seek and process health information.

Key aspects and tasks of Area 3

Amongst others, a key role will be played by the study of pre-knowledge, motivation, interests, cognitive biases, and expectations that influence the most suitable quantity, detail, context and presentation of information for specific consumers. We will define and inform the adaptation mechanisms needed to suit individual consumer profiles. Another key task is to design and run evaluation experiments on the proposed adaptive approaches.

Importance of cognitive psychology for A+CHIS

The area is led by the Institute of Psychology at the University of Graz. The institute conducts research into cognitive psychology, including such aspects as the identification of mechanisms of knowledge, motivation and learning capacity, that takes into account that human cognition is vulnerable to many known cognitive biases and misconceptions. As far as health information is concerned, this can result in problems such as over-information or over-diagnosis.

Role of the leading institute in Area 3

Research from Uni Graz will introduce new concepts relating to adaptive health information and help avoid such cognitive bias. We contribute to interactivity, adaptivity and personalization in knowledge assessment and instructional design.

Expected outcomes and contributions of Area 3

We build on existing instructional learning theories and combine this field of research with research on cognitive bias and misconceptions. Moreover, we use and enhance formative and summative evaluation methods to assess different degrees and forms of adaptive and interactive A+CHIS prototypes.