Limiting Attributes Prevents Learning
To keep up with the recognized 30% data churn rate in healthcare, at Meperia, we collect unlimited data attributes for any one product and from multiple sources. This allows us to continuously enrich the information, eliminate the bad data and provide insightful information needed to support clinicians and physicians at the point of care, throughout the supply chain and downstream, to accounts payable. Retaining attribute association from all sources creates a learning environment and allows each attribute to learn “what is” (or what an acceptable attribute can be), and more importantly, “what is not” in relation to a specific product. This allows us to systematically process data from multiple sources with differing states of data accuracy.
We store all input values and use “weight of evidence” (statistical analysis) to select the most appropriate values to form a single output attribute which can change over time based on changing input values of course. This is called natural non-judgmental data modeling, and this allows us to dynamically detect changes in a product attribute. Because we store all input values, we can go back in time to understand the impact and make the necessary adjustments at the point of impact. This is particularly useful when there are vendor mergers as clients catch up at different rates. And because we store all values, we can maintain both new and legacy values without detriment.
As more data is collected from multiple sources, relating to a specific product, new attributes emerge, and this creates a more complete understanding of the product and how it fits in a family of similar products. This is called Self-Revelation and Emergence and this allows Meperia to grow the understanding of a product and its related attributes over time.
At Meperia, we know that the collection of unlimited attributes creates a strong learning environment. Limiting attributes prevents learning and the point of all this is to thrive, not just survive. AI has to be used effectively today to keep up with the 30% data churn rate in the healthcare industry. For a greater understanding of the AI we leverage at Meperia read our whitepaper.
About the Author
Lee Ann McWhorter is the Vice President of Business Development at Meperia. Her primary goals are to reduce data error, reduce cost and improve patient safety in the healthcare industry. Lee Ann has worked with the founders of Meperia for over 16 years and understands the breakpoints in the healthcare supply chain and the need for unbiased, quality data moving forward. For more information about Meperia please reach out to email@example.com.