EdjAnalytics’ Predictive Inspection Model for FDA

Share

Under a contract with CDER, EdjAnalytics, a data science firm providing artificial intelligence, machine learning and advanced analytics for life sciences, has developed a predictive model to support the agency’s decision-making process to select manufacturing sites for inspection. “The model provides the likelihood of a facility having an inspection outcome of an OAI (Official Action Indicated), VAI (Voluntary Action Indicated) or NAI (No Action Indicated),” Edj says. “Additionally, the model produces information about what factors most influence a particular outcome.”

 

Edj began working on the inspection model 9/2018. “After completing the modeling earlier this month, Edj is currently working to assist the FDA in integrating the results of its analysis,” it says. Under the contract, Edj says it “trained machine learning models on historical inspection outcomes to create a predictive model. Features related to facilities, their associated products and their historical inspection outcomes were used as model inputs. Model development was guided by predictive error, implementation feasibility and the interpretability of findings to stakeholders.”

 

Read more