Accuracy rate >95% for the top-7 diagnoses produced
Sonioalong with Ecole Polytechnique has 1 submitted patent: “Real-time diagnosis aid method and decision-support system for medical diagnosis to a user of a medical system”
At the edge of research in fetal medicine and mathematics by helping to diagnose 294 syndromes based on 668 abnormalities visible on ultrasound.
Robust decision support system (DSS) with a success-rate >95% for top-7 diagnoses when the full phenotype is provided, using a large database of noisy real data. This makes it a software to help in the diagnosis of real cases in fetal medicine.
Leverages artificial intelligence to support clinical expertise and go beyond the capabilities of the human brain to provide unique clinical value. In this sense, it is able to manage in real time at 3200 possible combinations to guide towards the right diagnosis.
It takes into consideration multiple types of data : echographic visible signs, medical history, risk factors, and genetic information.
Provides satisfactory performances even on rare forms of rare diseases whose phenotype only partially corresponds to what is described in the literature. Indeed, in the fetopathology database used for validation, half of the anomalies found were atypical.
Sonio is continuously working to provide structured data for research. As of today, Sonio has put its expertise at the service of medical research by structuring a large-scale international database.
⁽¹⁾ Class I medical device, according to Directive 93/42 / EEC