Hevelius Report: Visualizing Web-Based Mobility Test Data For Clinical Decision and Learning Support

TL;DR

A visualization tool that abstracts six clinically relevant concepts from 32 features of Hevelius, and is an interactive app to meet the specific needs in different usage scenarios, that represents a promising solution for analyzing fine-motor test results and monitoring patients’ conditions and progressions.

Abstract

Hevelius, a web-based computer mouse test, measures arm movement and has been shown to accurately evaluate severity for patients with Parkinson's disease and ataxias. A Hevelius session produces 32 numeric features, which may be hard to interpret, especially in time-constrained clinical settings. This work aims to support clinicians (and other stakeholders) in interpreting and connecting Hevelius features to clinical concepts. Through an iterative design process, we developed a visualization tool (Hevelius Report) that (1) abstracts six clinically relevant concepts from 32 features, (2) visualizes patient test results, and compares them to results from healthy controls and other patients, and (3) is an interactive app to meet the specific needs in different usage scenarios. Then, we conducted a preliminary user study through an online interview with three clinicians who were not involved in the project. They expressed interest in using Hevelius Report, especially for identifying subtle changes in their patients' mobility that are hard to capture with existing clinical tests. Future work will integrate the visualization tool into the current clinical workflow of a neurology team and conduct systematic evaluations of the tool's usefulness, usability, and effectiveness. Hevelius Report represents a promising solution for analyzing fine-motor test results and monitoring patients' conditions and progressions.

Figures (10)

• Figure 1: Interactive visualization tool (Hevelius Report) developed to help users analyze mobility test results. After the user enters a patient's unique code, the tool provides a report of the patient's test results with (a) summary plots, (b) trajectory plots, (c) speed plots, and (d) additional information on the analyses. The interface is a single continuous page with a sidebar that allows users to jump to a particular section. The red asterisks (annotations in the figure; not present in the tool) indicate opportunities for user interactions.
• Figure 2: Factor analysis results. We performed factor analysis on the Hevelius test results from 247,667 healthy control participants and reduced the 32 Hevelius features to 6 factors. The coefficients representing the relationship between each factor and the features are displayed above. We mapped the 6 factors to 6 concepts: 1) deviation from straight line, 2) directional change from target, 3) pauses and jerks, 4) speed, 5) time inconsistency, and 6) speed inconsistency (Factors 1-6, respectively). Discussions with clinicians indicated that these concepts are clinically relevant.
• Figure 3: Example summary plot for a Parkinson's patient for a given timepoint and concept. The plot on the left compares the patient's result with that of the relevant patient and healthy control sub-populations. The plot on the right shows the progression of the patient's condition over time and compares the patient's progression with that of other patients.
• Figure 4: Example trajectory plots for a Parkinson's patient. The plot compares the mouse trajectory of the patient (blue) to example trajectories from healthy controls (gray) over two Hevelius sessions (two columns). Users can choose to see one average healthy control trajectory or a distribution of up to 20 trajectories. The $5th$, $50th$, and $95th$ percentile trajectories are selected based on a concept (in this example, the "deviation from straight line" concept).
• Figure 5: Example time-speed plots for a Parkinson's patient over two Hevelius sessions. The plot shows the speed of the mouse during the corresponding trajectories in the trajectory plot (Figure \ref{['fig:trajec-plot']}). Users can choose to see one average healthy control speed line (gray) or a distribution of up to 20 speed lines. The initiation phase (time between the start of the task and the start of mouse movement) is highlighted in the light blue shaded area.