Quantitative Decision Reliability of AI-Accelerated Computational Modeling and Simulation for Cardiovascular Devices

The FDA is seeking a contractor to develop methods for evaluating the reliability of AI-accelerated computational evidence in cardiovascular device regulatory submissions. The contractor will perform tasks such as reviewing surrogate modeling for cardiovascular device applications, developing surrogate models, and assessing decision reliability. The contract will be a firm fixed price type contract, and the period of performance is 12 months. The contractor must provide computational resources necessary for computational fluid dynamics simulations and surrogate model development. • The contractor will conduct a literature review of surrogate modeling methods used in cardiovascular devices applications. • The contractor will develop surrogate models of computational fluid dynamics and demonstrate a method for assessing if surrogate model errors could alter the final decision compared to the baseline simulation model. • The contractor will assess the likelihood that using the surrogate model in place of the reference computational fluid dynamics model leads to a different binary decision. • The contractor will develop a journal publication documenting all work performed under this contract.