Bialert Sepsis can predict a septic crisis in advance

Compilation
The Institute of Engineering Knowledge (IIC) has developed “Bialert Sepsis”the first Spanish artificial intelligence (AI) tool that predicts the risk of sepsis in 24 hours with a prediction accuracy of 96 percent.

BIAlert Sepsis anticipates what the doctor does not see and collects information from the patient every 30-45 minutes. history of the clinic the patient from the moment of his admission to the hospital. It also sends alerts to the medical team indicating the risk of sepsis 24 hours before the syndrome occurs. The solution was trained on data of more than 200,000 Spanish patients and the algorithm learns more and more over time.

Dr. Marcio Borges: “These are results that we cannot achieve either with any other technology or with the help of human potential, but with artificial intelligence we can”

Doctor Marcio Borgescoordinator of the multidisciplinary sepsis unit of the Son Llatzer University Hospital in Palma de Mallorca and president of the Sepsis Code Foundation, explained “BiAlert can see things that doctors don’t see because sometimes we don’t give importance to a certain piece of data that may have validity because we are limited in what technologies like artificial intelligence can do.” In addition, he emphasized Promote what is this tool? “These are results that we cannot achieve with any other technology or human potential, but with artificial intelligence we can.” Dr. Borges concluded.

In this way, the AI ​​warns about possible sepsis in a patient, indicates what tests should be performed on that patient and, finally, what type of treatment. care can be raised. “This is the future of not only warning about possible sepsis, but also guiding and helping doctors in making decisions. This is the support that these technologies will provide us. “It will not replace doctors, it will help us think”– Dr. Borges said in detail.

BiAlert has been running for two years at the Son Llatzer University Hospital in Palma de Mallorca.

Director of Healthcare at IIC, Eliza Martinexplained that they are implementing artificial intelligence software into their patient management systems at the 12 October Hospital in Madrid and the Albacete University Hospital Complex. This comes after the instrument worked for two and a half years at the Son Llatzer University Hospital (Palma de Mallorca).

For install BiAlert it is necessary to obtain approval from the hospital ethics committee and sign an agreement Data protection. You only need a server and a computer with some memory; and developing a vocabulary to standardize clinical criteria in each hospital. This does not create additional work for the professional. It is installed remotely and also tested remotely to ensure everything is working correctly.