NETANYA, Israel, February 20, 2017 /PRNewswire/ --
Intensix, developer of a real-time predictive analytics platform for early detection of patient deterioration in the ICU and high acuity departments of hospitals, announced today the positive results of a clinical study of the effectiveness of the Intensix platform in detecting and predicting sepsis in the ICU.
The pioneering Intensix platform is designed to apply machine learning to produce accurate detection and predictive clinical analytics for the ICU and high-acuity departments in hospitals. The analytics are intended to significantly improve clinical outcomes and reduce the average length of hospital stay, which in turn, reduces treatment costs, among other financial benefits.
The study was conducted by Brian Pickering M.D. and Vitaly Herasevich M.D., of the Mayo Clinic's Multidisciplinary Epidemiology Translational Research Intensive Care (METRIC) Group. The results were announced during a poster presentation at HIMSS17 titled Improving Sepsis Prediction by Advanced Model Development.
"The model developed by Intenix to detect sepsis in patients admitted to the ICU showed performance comparable to manual review. The developed model and system demonstrated high reliability with no down-time and ability to process flow of data needed for algorithm execution. Furthermore, the model detected other events that have signs similar to the physiological manifestation of sepsis. Additional studies that will better define the final model and evaluate the algorithm performance in multiple institutions are needed," said Dr. Herasevich.
Sepsis and septicemia are ranked among the top most costly in-hospital conditions in the US. Conventional management of sepsis, based on early goal-directed therapy, has not significantly decreased mortality in patients with septic shock. The study sought to investigate a possible solution to this.
The study included 782 patients, with a median age of 65 years. Median ICU length of stay was 35.5 hours. The Intensix platform showed a sensitivity of 90.5% and specificity of 88.5%. The Positive Predictive Value of the Intensix system was 71.5%. The Negative Predictive Value of the system was 96.7%.
"In the ICU, seconds count. Though a patient may look fine at one moment, the next moment he could be in crisis and seemingly without warning. The Intensix platform is designed to provide this warning, in enough time to avoid rapid patient deterioration and unfavorable outcomes. This is done by continuously monitoring patient parameters and collecting data from all available sources and instantly analyzing it. Our goal is to translate clinical intuition into a mathematical formula that can successfully produce timely alerts of pending crisis. This is an extremely complex mission, but one that for the first time, is now in the beginnings of being achieved by the Intensix platform," said Gal Salomon, CEO of Intensix.
Intensix provides healthcare providers and administrators with high-accuracy predictive analytics that improve clinical outcomes and reduce hospital costs. The Intensix innovative analytics solution detects deterioration in real-time and delivers predictive warnings during all phases of a patient's stay in the ICU and other high-acuity departments. Driven by innovative predictive modelling and advanced high-dimensional analytics techniques, the Intensix platform has the flexibility to manage entire patient populations as well as individualized treatment needs.
*Products and trade names mentioned herein are the trademarks of their respective owners.