Ge Precision Healthcare LLC

Karim Choukri

Uwe Wiedmann

Steven James Huff

Bulent Alpay

Harsha Aeron

A system comprising a processor configured to obtain a feature set for a component and generate random trees based on the feature set and a training data set, wherein each of the random trees can include at least one predictive value representing a probability of a feature of the random trees indicating a failure of the component within a period of time. The processor can also select a subset of the random trees based on the at least one predictive value, determine a likelihood of the failure of the component based on operational data for one or more devices and the subset of the random trees, and transmit an explanation to a remote device for the likelihood of failure of the component by indicating the feature selected from the subset of the random trees.

Growing community of inventors

Paris, France

Karim Choukri