Reinforcement Learning for Sampling on Temporal Medical Imaging Sequences
This work addresses the combinatorial complexity in sampling for medical imaging, offering a potential incremental improvement for dynamic image reconstruction in healthcare applications.
The paper tackles the problem of determining optimal sampling strategies for accelerated magnetic resonance imaging on temporal sequences by applying reinforcement learning algorithms (double deep Q-learning and REINFORCE) to learn sampling patterns, resulting in a proof of concept that these algorithms can discover effective patterns underlying a pre-trained reconstruction network.
Accelerated magnetic resonance imaging resorts to either Fourier-domain subsampling or better reconstruction algorithms to deal with fewer measurements while still generating medical images of high quality. Determining the optimal sampling strategy given a fixed reconstruction protocol often has combinatorial complexity. In this work, we apply double deep Q-learning and REINFORCE algorithms to learn the sampling strategy for dynamic image reconstruction. We consider the data in the format of time series, and the reconstruction method is a pre-trained autoencoder-typed neural network. We present a proof of concept that reinforcement learning algorithms are effective to discover the optimal sampling pattern which underlies the pre-trained reconstructor network (i.e., the dynamics in the environment). The code for replicating experiments can be found at https://github.com/zhishenhuang/RLsamp.