FDA Seeks Computational Algorithms for Adverse Events
FDA is seeking outside groups or individual data scientists for the development of computational algorithms to identify adverse event anomalies in open FDA adverse event data available on its openFDA platform. “Automated adverse event anomaly detection would augment the traditional data mining and case review approach by enabling the unsupervised identification of novel potential safety signals,” an FDA notice says. Under this “FDA Open Data Challenge,” the agency says such algorithms should detect anomalies automatically and without the use of known anomaly labeled training data.
The notice says that FDA staff experts will review submissions, which are due by 2/29. “Submissions will be judged based on the: impact to public health, potential to inform FDA post-market surveillance, novelty of the anomalies detected, reproducibility of the results, and generalizability of the algorithm,” it says. Some participants may also be invited to discuss their submission 3/27 at the agency’s Modernizing FDA’s Data Strategy public meeting that will focus on approaches to data quality, data stewardship, data exchange, and data analytics.