Security cameras can be very useful, but artificially intelligent cameras with cutting-edge image recognition algorithms? They can change our world. This was proven recently by researchers at Stanford University who used smart security cameras in a hospital to determine whether doctors, nurses and other hospital staff wash their hands as much as they should.
After developing the smart security camera, the team trained it on footage of people using the alcohol-based gel dispensers on two wards. There were 170 people -hospital staff – who entered the patient’s room at various times, but only 30 of them used the dispensers. This was used to teach the system the difference between a person using the gel and the person not using it.
After their training process was done, the smart security cameras were able to recognize with an accuracy level of 75% whether people were using the dispensers or not. Although 75% may seem not good enough, bear in mind that the human doing the same job was able to achieve 63% accuracy. So, it appears that machines win in this case.
As the team explains, the pilot project went well, so they’re now providing the smart cameras to three hospitals for one year to study whether or not they can have a positive impact on bugs spreading in hospitals.
Digital Trends (https://www.digitaltrends.com/cool-tech/ai-hospital-infection-handwashing/)