By Supriya Krishnan and Sian Claire Owen
What Tworek means, is that gaming technology has helped the team at the University of Calgary to create 97 virtual patients with ‘clinically accurate values and variety’.
Creating virtual patients can be challenging, especially regarding the time and effort taken in creating these resources, not only for the subject matter experts but also the developers.
“Regardless of the congruence of data categories, each VP content must be (re)-created and/or proofread independently, contributing to the development time,” she writes in Bio-Algorithms and Med-Systems (2010; Vol 6 (11):41-5).
Another challenge is ensuring the fidelity of the VP content. “In some VPs, users simply encounter the term ‘normal’ when the physical exam or laboratory finding is not abnormal,” she adds.
To counteract these problems, Tworek looked to gaming theory, and in particular to Dungeons & Dragons. In games such as Dungeons & Dragons, the characters move from one environment to another, and new characters need to be produced very quickly. This is achieved by randomising a range of values attached to characteristics, like weapons, height and eye colour.
In a clinical context, randomizing social history, laboratory and other investigation results “to default within the range of accepted physiological parameters for each finding” can produce a range of clinically accurate, and varied VPs. As a result, textual data was converted to numerical values. “In other words, the text strings were assigned to numbers,” said Tworek.
The process was outlined as follows:
- Create the VP template;
- Create data tables;
- Create specific data range by node;
- Randomly assign ID values in the data range to specific VPs;
- Create a report by linking randomized ID values to the text fields;
- Convert IDs to actual values and text;
- Proof read the VPs; and
- Upload content to the Open Labyrinth VP player.
“In the end, we produced a total of 97 VPs, each one taking 20 minutes to create,” she says.