Our technology has been vetted through fourteen peer-reviewed publications.

High accuracy, up to two days earlier

By incorporating patient vital sign and serum creatinine measurements, as well as
hourly changes in those measurements, Previse can predict acute kidney injury 48
hours before onset with 84% percent accuracy and a diagnostic odds ratio of 5.8.1

High accuracy, up to two days earlier
1. Mohamadlou H, Lynn-Palevsky A, Barton C, Chettipally U, Shieh L, Calvert J, Das R. Prediction of Acute Kidney Injury with
a Machine Learning Algorithm using Electronic Health Record Data. bioRxiv. 2017 Jan 1:223354. Read the paper here >

Uses minimal data, outperforms SOFA

Previse only requires values of each of heart rate, respiratory rate, temperature, serum creatinine, glasgow
coma score, and age. Previse demonstrated an area under the receiver operator characteristic (ROC) curve
(AUROC) of 0.841 at time of AKI onset while the sequential organ failure assessment (SOFA) score achieved
an AUROC of 0.762. Previse AUROC improved upon that of SOFA for all prediction windows (p < 0.01).2

2. Mohamadlou H, Lynn-Palevsky A, Barton C, Chettipally U, Shieh L, Calvert J, Das R. Prediction of Acute Kidney Injury with
a Machine Learning Algorithm using Electronic Health Record Data. bioRxiv. 2017 Jan 1:223354. Read the paper here >

Interested in learning more?

Contact Us