Cluster globally, Reduce locally: Scalable efficient dictionary compression for magnetic resonance fingerprinting

TL;DR

This paper introduces a scalable, efficient dictionary compression method tailored for magnetic resonance fingerprinting, leveraging Gaussian mixture models and incremental algorithms to handle large-scale, high-dimensional medical data with minimal information loss.

Contribution

It proposes a novel Gaussian mixture model-based compression technique with an incremental algorithm for large datasets, specifically applied to magnetic resonance fingerprinting.

Findings

Achieves 97% dictionary compression in MRI fingerprinting

Enables faster and more accurate map reconstructions

Handles high-dimensional data efficiently

Abstract

With the rapid advancements in medical data acquisition and production, increasingly richer representations exist to characterize medical information. However, such large-scale data do not usually meet computing resource constraints or algorithmic complexity, and can only be processed after compression or reduction, at the potential loss of information. In this work, we consider specific Gaussian mixture models (HD-GMM), tailored to deal with high dimensional data and to limit information loss by providing component-specific lower dimensional representations. We also design an incremental algorithm to compute such representations for large data sets, overcoming hardware limitations of standard methods. Our procedure is illustrated in a magnetic resonance fingerprinting study, where it achieves a 97% dictionary compression for faster and more accurate map reconstructions.