"Don't look at me, my hands are clean!"
Lack of hand hygiene is one of the leading causes in acquiring an infection. In spite of hospitals being always filled with alcohol-based gel dispensers and advisory posters, the fact of the matter is that it really isn't just effective.
A recent pilot study may just have found the solution.
The team in the Swiss Federal Institute of Technology (EPFL) in Lausanne tracked people around two hospital wards and automatically identified when they used gel dispensers using the combination of depth cameras and computer-vision algorithms. The success of the trial led to the group now going to fully kit out three hospitals to see if it puts a dent in the stubborn acquired infections statistics... for a whole year.
Initially, the group collected images from cameras installed overlooking corridors, patient rooms, and alcohol-based gel dispensers, among other places. Only 30 people correctly and used the gel dispensers, out of the 170 people they collected records from.
To detect the healthcare staff, track them as they move from one area to another across numerous cameras, and monitor their hand hygiene behaviour, the team used 80 percent of the images to train their algorithms. Then they tested the system on the remaining 20 percent, and achieved an accuracy of 75 percent in telling whether people had used the dispensers.
The SOP of the hospitals in this case have an individual assigned with a clipboard to have data collected. An assigned individual trying to do comparable monitoring had an accuracy of 63 percent across the same period. However, accuracy is only one factor. The assigned can only note the frequency of gel dispensers being correctly used, not provide a continuous map of everyone moving in and out of the areas 24/7.
But cameras can.
Alexandre Alahi at the Swiss Federal Institute of Technology (EPFL) in Lausanne says, "We're trying to shed light ont he dark spaces of healthcare. Understanding the problem is just the first step."
Philip Polgreen at the University of Iowa has used wearable technology in monitoring hand hygiene and identified data that wouldn’t have been possible with observers alone. For example, people were more likely to stick on to the correct procedures when other people are at present."We found that if you can improve the behaviour of a few very well connected people, you end up having a much bigger effect than when trying to target the overall average. This can tell us how to stop outbreaks,” he claims.
Clearly, there are privacy concerns since this would mean that the cameras would constantly monitor hospitals. But the depth cameras used were actually made to capture more information with regard to the position of an individual than their appearance. The resulting images are nothing more than human blobs.
“We can’t afford to have a doctor in a room 24/7, but we could afford an AI doctor every room, and every corridor too, leaving humans to do the most important jobs,” says Alahi.
Get weekly science updates in your inbox!