Inferring Visual Behaviour from User Interaction Data on a Medical Dashboard

Abstract

Making medical software easy to use and actionable is challenging due to the characteristics of the data (its size and complexity) and its context of use. This results in user interfaces with a highdensity of data that do not support optimal decision-making by clinicians. Anecdotal evidence indicates that clinicians demand the right amount of information to carry out their tasks. This suggests that adaptive user interfaces could be employed in order to cater for the information needs of the users and tackle information overload. Yet, since these information needs may vary, it is necessary first to identify and prioritise them, before implementing adaptations to the user interface. As gaze has long been known to be an indicator of interest, eye tracking allows us to unobtrusively observe where the users are looking, but it is not practical to use in a deployed system. Here, we address the question of whether we can infer visual behaviour on a medication safety dashboard through user interaction data. Our findings suggest that, there is indeed a relationship between the use of the mouse (in terms of clickstreams and mouse hovers) and visual behaviour in terms of cognitive load. We discuss the implications of this finding for the design of adaptive medical dashboards.

Modelling the interactive behaviour of users with a medication safety dashboard in a primary care setting

“…Table 1 shows the distribution of users and sessions per user group. (14) 32 (6) Other 4 (11) 22 (4) Following this, we used the Weka machine learning software [30] to classify the different types of users -primary and secondary -using machine learning algorithms, and used 10-fold cross-validation (CV) to evaluate the performance of the algorithms. In the sessions approach, since different session observations for the same user are not independent, we also performed an approximately stratified 3-fold CV, splitting the dataset in three folds of similar size and similar primary/secondary proportion but keeping all the sessions of each user in the same fold.…”

“…While implementing usability guidelines may address some of the most prominent and critical usability issues, users still feel overwhelmed by the amount of information on the screen. As a result, the interaction of clinicians with health data is not as smooth as expected, and existing evidence indicates that, in order to accomplish their tasks effectively, they demand 'just the right amount' of information [4].…”

“…A few examples highlight the benefits of the visual nature of dashboards, including successful interventions for diabetes care [2] and management of alerts triggered by drug-drug interactions [3]. Typically, medical dashboards display data in a tabular fashion and contain images, charts, numeric and textual information that often tax the perception and cognition of their users [4].…”

Long Term Energy Savings Through User Behavior Modeling in Smart Homes

“…Dashboard visual feedback is one of the best ways to help people interact with systems that generate large amounts of data, which would be difficult to comprehend without it [40]. An interactive dashboard can filter, summarize, and present information relevant to users by using simple widgets and charts.…”

Navigational Behavior Patterns of Learners on Dashboards Based on Assessment Analytics

“…For this reason, in this study, assessment analytics were handled as the data obtained from assessment tasks and students' interactions with feedback. While it is easy to track students' interaction with feedback through technological developments, few studies examined this situation (Kia et al, 2020;Yera et al, 2018). It is noteworthy that in these studies, the role of examinations was not the focus of any indicators.…”