EMERGENCY MEDICAL RESPONSE TRAINING – VIRTUAL REALITY

The EMRT-VR represents a groundbreaking training program inspired by my personal journey as an Emergency Medical Response (EMR) student. While the academic aspect of the EMR course was robust, the skills training test at the end left many of us certified with some uncertainties, highlighting the need for a more comprehensive and confidence-building training platform.

The concept of EMRT-VR aims to offer medical students and recertifying professionals an immersive learning experience that transcends theoretical knowledge. This platform enables trainees to engage in realistic virtual emergency scenarios, allowing for hands-on practice in primary medical assessment. Trainees, immersed in a virtual emergency scenario, interact with a 911 caller-avatar, utilizing the procedural SAMPLE* questioning to triage or treat the simulated emergency.

SAMPLE stands for:

  • S: Signs and symptoms
  • A: Allergies M: Medications
  • P: Past medical history
  • L: Last oral intake E: Events leading to the illness or injury
  • E: Events leading to the illness or injury 

This PDF is not viewable on a phone device. Please use a high bandwidth connection to view or have alot of patience.

After numerous rounds of trial and error, we successfully achieved a flawlessly functioning prototype, provided there was a strong and uninterrupted internet connection. While this video may appear ‘canned’, our dedicated efforts are evident in the following videos, showcasing a smoother and more real-time interaction between the trainer and Albert, our 911 patient. We developed a straightforward rule-based model called Albert’s Decision Tree, in collaboration with a medical response subject matter expert. For this prototype, we explored and utilized affordable and user-friendly tools such as speech-to-text and text-to-speech options (Watson vs. GCP), as well as Unity or UE4. Should this concept receive funding for further development, we intend to incorporate LLM models.

A simple rule-based model referred to Albert’s Decision Tree that was developed with a medical response SME. For this prototype, we used explored and used simple, affordable tools – speech-to-text/text-to-speech options (Watson vs. GCP), Unity or UE4.  Further development, would involved LLM models if this concept were funded. 

Haptic technology is here and waiting for it to be used in simulation training.

This video hints at what could be possible given the financial resources and a larger development team.