Prescriptive Analytics for Preventable Harm

The Jvion Machine

Welcome To Eigen-enabled Prescriptive Analytics

The Jvion Machine is built using an Eigen Sphere engine—an approach that is a proven asset in solving complex challenges such as quantum mechanics, search and consumer behavior, facial recognition, and now patient deterioration.
This Eigen-based approach is used to enable a comprehensive patient view that is amplified beyond the risk of an event to the clinical actions that will improve outcomes and drive patient engagement. The breadth of application and specificity of the recommendations delivered by the machine are empowering providers with a solution that can extend to all aspects of patient care across ambulatory and inpatient settings.
To understand how Eigen-based machines more effectively drive patient impact while helping providers lower costs, we have to understand the underpinnings of the Eigen Sphere engine and its key differentiation from simple predictive analytic methods including neural networks, random forest, and regression analysis machine learning models.

Harnessing The Power of Eigen Spheres

The Eigen Sphere engine is an n-dimensional space upon which millions of patients are mapped against tens-of-thousands of Eigen Spheres. Each Eigen Sphere comprises patients who clinically and/or behaviorally demonstrate similarities. These similarities impact patient physiology and engagement propensity. Hence, they impact the effectiveness of interventions and clinical actions.
The machine is perpetually adjusting the spheres and patients mapped within the engine with every new piece of information entered into the machine. This continual fine-tuning enables the localization and ongoing performance improvements that are characteristic of prescriptive analytic technology.
  • 4 Keys to Successful Adoption of the Cognitive Machine and Artificial Intelligence in Healthcare
    Technology is the means, not the end. The cognitive machine for healthcare— a revolutionary “brain” for clinical decisio...
    10 Nov 2017 | Read More
  • Strategic Considerations for Adopting Clinical AI
    It may be late compared to other industries, but Artificial Intelligence (AI) is now moving rapidly into healthcare, repre...
    10 Dec 2019 | Read More
  • Making Healthcare IT Heroes: Cognitive Technology, Clinical Precision and IT Harmony
    Being a healthcare IT leader can be a thankless job. A CIO at a forum a couple of years back described his health system...
    28 Feb 2018 | Read More
Stay Informed
  • Jvion Featured as Leading Player of Machine Learning in Medicine Global Market
    18 Apr 2019 | Read More
  • Jvion's Dr. John Showalter Highlighted for Bringing AI to Veterans
    09 Jul 2018 | Read More

Embodying Risk and Effective Action in Clinical Vectors

Each patient within the Eigen Sphere engine is moving at a certain speed and trajectory toward or away from risk; this is called a vector. The clinical value of the machine is derived by its ability to see beyond just the patients who are high-risk to account for those individuals who are on a trajectory toward risk. The goal is to change the speed and trajectory of this movement by applying interventions. The machine models the impact that each intervention will have on a specific patient’s risk and provides recommendations based on each intervention’s performance.
For example, the machine can identify the individuals at high-risk of a pressure injury and the patients who are on track to become high-risk if an action is not taken to change the outcome. The actions that will change a trajectory and improve quality are modeled and communicated back to the caregiver though prioritized recommendations. Once the actions are taken, the machine accounts for the intervention and continues to monitor a patient’s movement relative to the pressure injury risk.
The power of the machine comes from its ability determine a patient’s trajectory toward all target medical events, at the same time, and perpetually. More than a quadrillion (1015) considerations are made for each patient within the machine. But, the risks and recommended actions delivered to a clinician are presented in one, cohesive output. This is the real hallmark of the Jvion machine: its the capability to “think” about patients in the same way as a clinician. The machine views every patient as complex, constantly changing individuals. This  patient view translates into an ability to effectively reduce a clinician's cognitive load by providing the information that will direct care to the right patients at the right time with the right clinical actions.
The major advantages of this technique over any other regression model or a standalone neural network are multifaceted:

Personalized Interventions

With this technique, we can not only find high-risk patients but can extend the effective view of the machine to identify which sphere will or will not work with certain interventions. The degree of specificity enabled by the machine pin-points the best individualized clinical action resulting in an intervention tailored to the specific factors, risks, and engagement propensity attributed to an individual patient

Impact and Effectiveness

Simply knowing which patient is high-risk is pointless if a provider cannot make an impact on the outcome. The Eigen Sphere technique allows providers to expand the concept of intervention effectiveness to know which patients are not just high or medium risk but—more importantly—where a provider can impact a patient’s outcome

Patient-centered Self Learning

This technique inherently incorporates many aspects of self-learning evident in advanced prescriptive analytic systems. The holistic patient model incorporates and continuously adjusts the Eigen Spheres based on clinical, socio-economic and behavioral data from millions of outcomes. This results in a more complete self-learning prescriptive machine, which is simply not possible in a typical regression model