AI-SCoRE stands for Artificial Intelligence – Sars Covid Risk Evaluation.
The revolutionary service was developed thanks to the collaboration between San Raffaele, Microsoft and NVIDIA and has a twofold objective: on the one hand to identify the subjects to be protected more, i.e. people at greater risk of developing serious forms of Covid-19 if infected by the virus; on the other hand to recognize among patients who show symptoms of the disease, those who will have the worst prognosis.
It will work thanks to an AI algorithm that will integrate diagnostic images, clinical and laboratory parameters, inflammatory status, and genetic profile of the patient and the virus.
How AI-SCoRE works
It is an autonomous learning platform which, on the basis of a set of indicators, calculates the possibility of developing serious forms of COVID-19 for an individual, thus enabling the optimisation of health intervention.
“In Phase 1 we could not identify in advance the most fragile people among the patients with the first symptoms of the disease – explained Carlo Tacchetti, coordinator of the project – We want to be able to do it precisely and quickly, because only in this way we will be able to understand who are the subjects who, once infected, need timely treatment, even in the absence of serious symptoms. Obviously, our dream is to push beyond this potential and take advantage of this opportunity to develop transversal algorithms able to identify the subjects most at risk even in the general population, and not only in subjects with COVID-19 suspicion“.
The functioning of the platform is guaranteed by Microsoft’s technology for data analysis and will allow, thanks to artificial intelligence, to collect, process and manage different data from multiple sources, so as to provide the necessary information to medical staff in a timely manner.
“At Microsoft, we help organizations do more with technology, and this also applies to the healthcare world,” said Veronica Jagher, Director of Healthcare Market Microsoft Area Western Europe. “Our primary goal is to use Artificial Intelligence systems to analyze large amounts of heterogeneous data and understand the factors that determine everyone’s health, thereby accelerating the implementation of personalized and precision medicine. AI-SCoRE goes exactly in this direction,” she added.
Where is the project at
The collection of data on patients to make AI-SCoRE work has already started: subjects have been identified between San Raffaele Hospital, Bolognini Hospital of Seriate and Monzino Cardiological Centre.
Thanks to this database, the algorithm will be “trained” on the dedicated software infrastructure, to learn how to combine data in the right way and indicate the risk of the individual patient to incur serious complications due to the disease.
Following the algorithm training, result validation tests will be performed on a second sample of patients and in any prospective studies.