MedAware’s medication surveillance technology identifies ADEs and eradicates catastrophic medication errors by applying advanced machine-learning algorithms and outlier detection mechanisms similar to fraud detection solutions in use by financial institutions worldwide. By continuously mining data gathered via millions of EHRs, the software is able to flag potentially lethal prescriptions that… · Moret are in conflict with the profile of the patient, physician, or institution. In addition, MedAware actively monitors each patient to identify and warn of situations in which changes in a patient’s diagnostic results renders one of his/her active medications a dangerous outlier. These difficult or nearly impossible to anticipate errors would otherwise go undetected by current rule-based solutions. The company’s unique, real-time approach to identifying ADEs and preventing medication errors saves lives, improves patient safety and outcomes, and significantly reduces avoidable risks and costs.
May 2012—Technology to eliminate prescription errors
MedAware provides an innovative solution to the need for detecting and eliminating prescription errors. MedAware’s patent-pending technology uses big data… · More analytics and machine learning algorithms to analyze large-scale data of Electronic Medical Records (EMRs). By deploying MedAware’s proprietary algorithms to mine the data gathered via millions of EMRs, MedAware’s engine builds a mathematical model which represents real-world treatment patterns. A prescription largely deviating from the standard treatment spectrum is likely to be erroneous.
The underlying assumption behind this solution is that most physicians perform well most of the time. Consequently, prescription patterns of thousands of physicians treating millions of patients can be used to determine the “normal” treatment spectrum. A prescription largely deviating from this spectrum is likely to be erroneous. One can find many similarities between this approach and the one successfully employed for fraud detection at the point of sale. Both apply cutting-edge big data analytic capabilities to identify outliers from a trend or practice in order to identify suspicious or erroneous transactions.