The Althaia Foundation has an innovation and development project underway to prevent falls in frail admitted patients, using artificial intelligence. The study, which is being carried out in collaboration with the company Robbie AI, makes it easier for healthcare professionals to foresee any fall hazard situation and act to prevent it from happening. In this way, safety is increased for patients who, either due to age or functional capacity, are considered at-risk. The results of the pilot study carried out to evaluate its effectiveness are very positive, with falls reduced by 90% overall, and zero falls in these units during the last three months.
Robbie AI is a fall prevention technology that uses real-time computer vision. The solution, which is fully automated, continuously analyzes patients’ body movements and facial expressions to predict risk situations and prevent patients from falling. When a risk of falling is detected, the system sends customizable alerts to the nursing team, with enough time to prevent the fall.
For two years, a pilot study has been carried out in the Acute Geriatrics Unit of the Sant Joan de Déu Hospital in Manresa and in the Long Stay Unit of the Hospital Center, to adjust the Robbie AI workflow to the needs of the hospital’s nursing professionals and to evaluate the effectiveness of the system. It has been a phase of adaptation of the artificial intelligence algorithm that has been trained according to the different types of alerts indicated by the healthcare team.
This implementation of artificial intelligence at the Althaia Foundation exemplifies how technology can be successfully used to improve patient safety in hospital settings, and can serve as a model for other health and social institutions.
100% reliable alerts
During the pilot study, an average of 16 alerts were sent per day/bed and, for high-risk patients, an average of 26 alerts per day/bed, with 100% reliability and less than 1% false positives. The alerts indicate different actions that the patient performs that can pose a danger, from getting out of bed and walking erratically, to difficulties sitting in bed or in the chair, among others.
The results of the pilot study show a 90% reduction in fall situations, with an average of 60 seconds in response time and 100 meters of distance between professionals and the patient.
It should be noted that the system had been trained and tested with more than 2 billion images and 10,000 patients from different centers. Robbie AI’s architecture, deployed on Nvidia Jetson’s Edge Artificial Intelligence platform, allows the removal of patient images and data to ensure maximum privacy.
In this pilot, “one of the most conclusive results has been the low number of false positives. Robbie AI is the only solution that intelligently distinguishes a real risk of falling, and this is crucial for the nursing team that manages many daily alerts,” says Bartomeu Ayala, head of Social and Health Innovation at the Althaia Foundation and member of the Steering Committee of the Health and Artificial Intelligence Program of the Generalitat de Catalunya.
“Althaia’s commitment to patient safety makes them an ideal partner for Robbie AI,” says Karen Marquez, CEO and co-founder of Robbie AI. “Their continued efforts to refine fall prevention workflows with Robbie AI’s solution have allowed us to achieve results that can be extended to other centers.”
SOURCE: Businesswire
