Abstract
Most diagnostic AI-algorithms are in need of high-quality observations to produce reliable results. The same is true for any human being, who is performing diagnostic reasoning. It is impossible to state an accurate and reliable diagnosis using false or faulty observations. In this paper an architecture to support the diagnostic algorithms based on multilevel flow models is presented. However, any diagnostic algorithm would benefit from using this architecture. The diagnostic algorithms based on multilevel low models already exist. This paper outlines the requirement for a system that shields the diagnostic algorithms from the uncertainties in the process.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
