JMIR Publications

ABSTRACT

Background:

Shortage of human resources, increasing educational costs and need to keep social distances in response to the COVID-19 worldwide outbreak prompt the necessity to move from a in-presence, bed-sided, didactic sessions towards novel tele-health teaching. Virtual Patient Simulators may partially accomplish these needs.

Objective:

To introduce the underlying rationale and most important features of Hepius, a Cognitive Tutor and Simulator, mimicking the doctor’s diagnostic process. To provide preliminary results on undergraduate students short-term knowledge change after Hepius use.

Methods:

Hepius was built up to let the student gather clinical information from input scenario, patient’s medical history, physical examination and tests results and address these information by prompting hypothesis generation, binary and pattern analyses. Natural language processing (NLP) techniques were used to simulate daily life student’s interaction with a real patient. Fifteen students attending the 5th year of the Humanitas University Medical School underwent two identical multiple choice question tests before and after performing a pulmonary embolism simulation with Hepius. Multiple choice test was made up of 22 questions, 11 of which (core questions) were specifically designed to evaluate knowledge acquired directly from performing the simulation.

Results:

Mean post-simulation test score was greater (P=0.0001) than pre-simulation score. There was an improvement (P=0.0004) in mean core questions score from pre- to post-simulation. Correct responses to core questions were greater (P=0.0027) post- compared to pre-simulation.

Conclusions:

The use of a software based on NLP techniques, specifically designed to mirror doctor’s diagnostic process, proved to be effective in promoting the short-term enhancement of students’ knowledge by a tele-health teaching approach.