The capabilities enabled by Jvion’s Cognitive Clinical Success Machine are highlighting the massive divide between model-based, predictive functionality offered within Electronic Health Record (EHR) solutions and robust, advanced Eigen-sphere based cognitive machines.
Atlanta, GA, July 25, 2017 --With the anticipated big-bang roll out of predictive/machine learning (sometimes referred to as Artificial Intelligence) capabilities by the major Electronic Health Record (EHR) vendors, there will be a sustained increase in market rhetoric focused on the intelligence of these new models. As details emerge on these solutions, the differences in capabilities between model-centered, EHR-embedded predictive functionality and patient-centered, true cognitive machines are coming into focus. This differentiation is especially true for Eigen-sphere based solutions like Jvion’s Cognitive Clinical Success Machine.
Current EHR-embedded solutions provide a menu of models designed to predict some aspect of patient risk. Most are use-case driven whereby each model is tuned to a specific illness and is measured on its ability to identify high-risk patients. The risk predictions are touted as accurate and contextually aware—information about a use case is presented within the patient record. They are baked into the care episode as point-in-time risk predictions, which are rendered at specific junctures in the continuum from perioperative, post-op, through post-acute.
“The challenge with current EHR-embedded predictive solutions is that they provide very narrow views into patient risk and focus primarily on the top 5-20% of the patient population that falls into the high-risk bands,” explained Dr. John Showalter, Chief Product Officer for Jvion. “The limited applicability of the solutions makes them hard to action. Oftentimes the people identified as at-risk are so complex and ill that we cannot do much to change the outcome. And the models driving these insights vulnerable to overfitting. They are limited in effectiveness and applicability, and they lack insights into the recommended actions that will best mitigate risk and drive patient engagement.”
Eigen-sphere based cognitive machines like Jvion’s Cognitive Clinical Success Machine are designed to render an ultra-definition patient view that delivers a multidimensional and longitudinal window into patient risk and the recommended actions that will improve outcomes. They produce an Eigen propensity biography for each patient that accounts for all aspects of disease etiology including the clinical, behavioral, and socioeconomic determinants of health. Disease and deterioration are assessed within the context of the entire patient and his/her environment to determine the cognitive propensity measure for a target illness or group of conditions. The Eigen architecture extends to deliver the recommended actions that will reduce the likelihood of an adverse medical event based on a patient’s likelihood to engage and potential barriers to intervention success.
Showalter continued, “With Eigen-sphere based cognitive machines we are creating a clinical appliance that ‘thinks’ about every patient in the same way as a clinician—as a complex, constantly changing individual impacted by thousands of internal and external factors. The idea is to deliver a machine that reduces the cognitive load of our caregivers by better and more precisely identifying the people who are at risk where we can have the biggest impact on health outcomes. EHRs are designed to help facilitate data gathering, standardization, and reporting. Cognitive machines are designed to drive more effective patient care.”
Jvion, Inc. (Jvion) delivers a Cognitive Clinical Success Machine that serves as a high-performance appliance for providers and the healthcare community. It activates macro and micro-level recommendations that help healthcare providers who need ultra-definition patient-level prioritizations, clinical action insights, and suggestions produced with unmatched speed, clinical applicability, and verity. The machine delivers the action-level recommendations that will best reduce the likelihood of an adverse event. This capability is enabled by a cognitive engine driven by horsepower that is based on more than a quadrillion clinical and non-clinical considerations and thousands of data elements. The machine’s thousands of self-learning Eigen spheres are applied to this data for each patient in real time to render an Eigen Propensity Biography that delivers a view into a patient’s total health 30,60,90, up to 365 days in the future. This machine is helping hundreds of hospitals across the nation reduce target illnesses and diseases.
One of the reasons Jvion’s solution is independently ranked number one in clinical predictive science is because the machine is more than accurate, it is effective. Our approach mitigates the “accuracy fallacy” perpetuated within the industry by delivering a true picture of individual patient risk along with adjacent risks and actions that will lead to better health outcomes. Because Jvion’s machine works as a cognitive appliance, it plugs in directly to the existing Electronic Medical Record/clinical systems to deliver recommendations seamlessly into the organic workflow. Clinician and caregiver adoption of Jvion’s recommendations is accelerated because of the “on-demand” nature of the information. The machine outperforms and outsmarts even the highest performing predictive solutions/approaches available. And this performance hasn’t gone unnoticed; Jvion’s solution has won numerous external awards including designation as the #1 Predictive Provider in Healthcare by Black Book Market Research.