Small Scale Test of the INGENIOUS triage and facial recognition applications (SST#6)
When responding to mass casualty incidents all First Responders are faced with a number of competing challenges and responsibilities to take control of the situation and deal effectively and efficiently with the wounded and deceased.
Primary Triage is designed to identify the health status of the patients at the scene to allow healthcare providers to administer treatment to the most serious cases. The triage officer, who doesn’t necessarily need to be a medic, sets up a fast labelling system to allow others to complete medical intervention or evacuation to hospital.
Triage uses a simple colour coded system to grade the condition of the patient so resources are deployed to the right person at the right time.
INGENIOUS provides a Triage device to attach to the victims with a corresponding app for any Android mobile phone to easily:
- update the details of the patient and manually set the triage colour
- link patients under a unique incident number
- follow patient transfer via GPS locations
- update the patient status on scene if their condition changes
Test 1 – Triage App
The partners of INGENIOUS from the Police Service of Northern Ireland invited members of their Disaster Victim Identification Team, (DVI) and representatives of the Northern Ireland Fire Brigade and Red Cross to a simulation of a building collapse resulting in a number of casualties and fatalities, which took place during the Small Scale Test #6 (SST#6) of INGENIOUS on November 2, 2021. We constructed a preloaded database of 45 fictitious employees’ biometric data from a Computer tech company and then gave the DVI Team 8 patient profiles to upload onto the system and activate the device. The patient information was loaded via NFC (Near-field Communication) by tapping the phone onto the device, in a similar manner to a contactless payment transaction.
The team were very impressed at:
- the speed at which the information was uploaded allowing the team more time to carry out other duties
- by the small size and weight, (18 grams) of the triage box meaning it would be unlikely to get in the way of medical interventions or impede transportation of the victim.
The triage box can either be attached to the patient via a locking bracelet or lanyard depending on injuries. This was important to ensure the unit was not lost during transportation.
The LED light emits a powerful coloured light associated with each triage colour. This is really important so it can be easily seen from a distance and in poor light/smoky environments.
Test 2 – Face Recognition App
One of the biggest challenges at a major incident is the fast identification of the victims as anxious friends and family wait for news of their loved ones. Facial recognition (FR) software is not new but the INGENIOUS solution work is ongoing on identifying patients who may have facial injuries or their faces may be obscured with blood or grime. The system was tested on both counts and showed promising results.
The picture below shows an injured face with a 65.84% match with a known dataset. The project will continue to improve on this percentage but it must be pointed out that this is only a step in the full identification process. Human oversight will always take precedence
As Facial Recognition software is a hot topic for individual privacy rights debate, discussions took place about the GDPR and ethics with the project’s ethics advisor and other partners. This area will continue to be explored throughout the lifetime of the project.
- The GPS component to track patient movement will be tested in the next iteration of the triage app
- work will be completed on improving the identification percentages of injured faces on the Facial recognition app
- Ethical and legal guidelines will continue to be explored by the project members and Ethic’s Advisory Board.