Cover Tree Compressed Sensing for Fast MR Fingerprint Recovery

Mohammad Golbabaee; Zhouye Chen; Yves Wiaux; Mike E. Davies

Cover Tree Compressed Sensing for Fast MR Fingerprint Recovery

Mohammad Golbabaee, Zhouye Chen, Yves Wiaux, Mike E. Davies

School of Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We adopt data structure in the form of cover trees and iteratively apply approximate nearest neighbour (ANN) searches for fast compressed sensing reconstruction of signals living on discrete smooth manifolds. Levering on the recent stability results for the inexact Iterative Projected Gradient (IPG) algorithm and by using the cover tree's ANN searches, we decrease the projection cost of the IPG algorithm to be logarithmically growing with data population for low dimensional smooth manifolds. We apply our results to quantitative MRI compressed sensing and in particular within the Magnetic Resonance Fingerprinting (MRF) framework. For a similar (or sometimes better) reconstruction accuracy, we report 2-3 orders of magnitude reduction in computations compared to the standard iterative method which uses brute-force searches.

Original language	English
Title of host publication	Proceedings of the 2017 IEEE Workshop on Machine Learning for Signal Processing (MLSP)
Publication status	Published - 1 Jun 2017

Keywords / Materials (for Non-textual outputs)

Statistics - Machine Learning

Access to Document

pdf

http://adsabs.harvard.edu/abs/2017arXiv170607834G

Cite this

@inproceedings{2190193b8b5e4ba2b06e0bd9f47e8164,

title = "Cover Tree Compressed Sensing for Fast MR Fingerprint Recovery",

abstract = "We adopt data structure in the form of cover trees and iteratively apply approximate nearest neighbour (ANN) searches for fast compressed sensing reconstruction of signals living on discrete smooth manifolds. Levering on the recent stability results for the inexact Iterative Projected Gradient (IPG) algorithm and by using the cover tree's ANN searches, we decrease the projection cost of the IPG algorithm to be logarithmically growing with data population for low dimensional smooth manifolds. We apply our results to quantitative MRI compressed sensing and in particular within the Magnetic Resonance Fingerprinting (MRF) framework. For a similar (or sometimes better) reconstruction accuracy, we report 2-3 orders of magnitude reduction in computations compared to the standard iterative method which uses brute-force searches.",

keywords = "Statistics - Machine Learning",

author = "Mohammad Golbabaee and Zhouye Chen and Yves Wiaux and Davies, {Mike E.}",

year = "2017",

month = jun,

day = "1",

language = "English",

booktitle = "Proceedings of the 2017 IEEE Workshop on Machine Learning for Signal Processing (MLSP)",

}

TY - GEN

T1 - Cover Tree Compressed Sensing for Fast MR Fingerprint Recovery

AU - Golbabaee, Mohammad

AU - Chen, Zhouye

AU - Wiaux, Yves

AU - Davies, Mike E.

PY - 2017/6/1

Y1 - 2017/6/1

N2 - We adopt data structure in the form of cover trees and iteratively apply approximate nearest neighbour (ANN) searches for fast compressed sensing reconstruction of signals living on discrete smooth manifolds. Levering on the recent stability results for the inexact Iterative Projected Gradient (IPG) algorithm and by using the cover tree's ANN searches, we decrease the projection cost of the IPG algorithm to be logarithmically growing with data population for low dimensional smooth manifolds. We apply our results to quantitative MRI compressed sensing and in particular within the Magnetic Resonance Fingerprinting (MRF) framework. For a similar (or sometimes better) reconstruction accuracy, we report 2-3 orders of magnitude reduction in computations compared to the standard iterative method which uses brute-force searches.

AB - We adopt data structure in the form of cover trees and iteratively apply approximate nearest neighbour (ANN) searches for fast compressed sensing reconstruction of signals living on discrete smooth manifolds. Levering on the recent stability results for the inexact Iterative Projected Gradient (IPG) algorithm and by using the cover tree's ANN searches, we decrease the projection cost of the IPG algorithm to be logarithmically growing with data population for low dimensional smooth manifolds. We apply our results to quantitative MRI compressed sensing and in particular within the Magnetic Resonance Fingerprinting (MRF) framework. For a similar (or sometimes better) reconstruction accuracy, we report 2-3 orders of magnitude reduction in computations compared to the standard iterative method which uses brute-force searches.

KW - Statistics - Machine Learning

M3 - Conference contribution

BT - Proceedings of the 2017 IEEE Workshop on Machine Learning for Signal Processing (MLSP)

ER -