On the Information Theory for Magnetic Resonance Imaging

This work presents the mathematical formulation of the Magnetic Resonance Imaging (MRI) system modeled as a wireless communication system to establish its information theory foundations. The MRI system conceived as a source-sink communication system has channel impairments that affect the transmitted data. The information source is a stochastic process that produces a sequence of information symbols governed by a set of probabilities. The adverse effects on the transmitted MRI signal shall limit the amount of information capable of being received at the sink. Therefore, reliable detection at the receiver shall be accomplished by estimating the channel capacity and an approximation of the source entropy. Modeling the MRI system using a wireless model shall simplify the receiver architecture, yielding new methods to improve MRI signal acquisition, i.e., different values of bandwidth and signal strength yield the same channel capacity. Achieving capacity bridges information and computation efficiency.Clinical relevance This work establishes the basics to reduce MRI scan times by acquiring sufficient information with less redundant information.