Geometry of Deep Learning for Magnetic Resonance Fingerprinting

Mohammad Golbabaee, Dongdong Chen, Pedro A. Gómez, Marion I. Menzel, Mike Davies

Research output: Contribution to journalConference articlepeer-review

21 Citations (SciVal)


Current popular methods for Magnetic Resonance Fingerprint (MRF) recovery are bottlenecked by the heavy storage and computation requirements of a dictionary-matching (DM) step due to the growing size and complexity of the fingerprint dictionaries in multi-parametric quantitative MRI applications. In this paper we study a deep learning approach to address these shortcomings. Coupled with a dimensionality reduction first layer, the proposed MRF-Net is able to reconstruct quantitative maps by saving more than 60 times in memory and computations required for a DM baseline. Fine-grid manifold enumeration i.e. the MRF dictionary is only used for training the network and not during image reconstruction. We show that the MRF-Net provides a piece-wise affine approximation to the Bloch response manifold projection and that rather than memorizing the dictionary, the network efficiently clusters this manifold and learns a set of hierarchical matched-filters for affine regression of the NMR characteristics in each segment.
Original languageEnglish
Pages (from-to)7825-7829
Number of pages5
JournalIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publication statusPublished - 1 May 2019


  • Magnetic resonance fingerprinting
  • deep learning
  • dictionary
  • inverse problem
  • manifold compressed sensing

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Geometry of Deep Learning for Magnetic Resonance Fingerprinting'. Together they form a unique fingerprint.

Cite this