ExplainerAI™ - Foster Trust with Clinicians and Patients
ExplainerAI™ is the first-of-its-kind platform that ensures compliance with your AI Governance standards while preventing model drift and bias through deep analytics across your entire portfolio of AI and ML models. Built to foster clinician adoption through trust and transparency, ExplainerAI™ is designed on the four pillars of NIH’s AI Governance Framework: trustworthiness, fairness, transparency and accountability.
Predictive risk factors, algorithm configuration parameters, performance analytics are presented in real-time at the patient and population level.
ExplainerAI™ dashboards are where all measures are aggregated and display model performance at-a-glance, enabling users to effectively identify patient care gaps for each measure and act accordingly. Users can drill down to better understand model configuration, at very granular levels, and performance over time.
ExplainerAI™ dashboards aggregate model data and breaks it down by demographic, socio-graphic and other key measures to ensure that there are no biases towards race, gender, economic group or other cohort that may be inadvertently disadvantaged by the model or its operational implementation.
Explainer AI™ can be configured to map to an organization’s AI governance requirements for explainability.
Demographic Analysis
Breakdown your model performance into its demographic components to ensure you do not have any inherent bias built into your model.
Model Performance vs Actual Outcome
Measure your models performance vs. actual outcomes and analyize your outcomes by demographic. This enables you to understand both the model performance and to see if the bias is the result of actual results or model error.
Transparancy
Enable everyone to understand why a model is making a decision about a particular patient. This is a crucial component when it comes to building trust and understanding in the front line staff.
Additional Features + Benefits
-
Works with any model
Whether you are using a Cognome model, one built by your staff or one purchased by a third party, ExplainerAI™ can integrate into any of them and provide the transparency your organization needs.
-
Monitor Model Drift
See model drift as it happens and re-train your model so that it continues to perform at the top of its game.
-
Everything in one place
See how all of your models perform in one place with one user interface. Don't force your staff to look for model performance in a mosaic of different locations and interfaces.
-
Governance
ExplainerAI is a core component of deploying any AI responsibly. Allow everyone to understand how models are performing and ensure that no portion of your population is being under served.
-
Built to Open Standards
Built to foster adoption through trust, ExplainerAI™ is designed on the four pillars of NIH’s AI Governance Framework: trustworthiness, fairness, transparency and accountability
-
Extendable to the product of your choice
Do you already have a dashboard you like to use? No problem. ExplainerAI™ is extendable to Power BI, Tableau and other BI platforms.