Matrix factorization from non-linear projections: application in estimating T2 maps from few echoes

Khurana, Prerna; Bhattacharjee, Protim; Majumdar, Angshul

doi:10.1016/j.mri.2015.04.007

Cited by 2 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is an analysis prior sparse recovery problem from non-linear measurements. There is no algorithm to solve (9). In the next section we derive a simple algorithm for the same.…”

Section: T1 Map Estimationmentioning

confidence: 99%

“…In this work, we propose to sub-sample the K-space for both PD and T1 estimation. This has not been attempted in [9]; there in, it is assumed that the image is generated from the fully sampled K-space. Our objective is to have a scan time comparable to that of an ordinary scan, i.e.…”

Section: Experiments On T1 Map Estimationmentioning

confidence: 99%

“…We follow an experimental protocol similar to the previous work [9]. We assume a saturation recovery spin echo sequence; there are other fast mapping techniques based on pulse engineering and sequence designing, but we are not discussing them in this study.…”

Section: Experiments On T1 Map Estimationmentioning

confidence: 99%

“…In this work we propose a technique to reduce the scan time to the same order of magnitude as an ordinary single echo MRI scan. Recently [9] a matrix factorization based method was proposed to address this problem. It modeled the quantitative map as a lowrank matrix.…”

Section: Introductionmentioning

confidence: 99%

“…We assume PD and T1 maps to be sparse in some transform domain. There are two advantages of our proposed approach over the prior study [9]: 1. In MRI reconstruction the sparse model is more well known and well understood compared to the low-rank model.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Fast Acquisition for Quantitative MRI Maps: Sparse Recovery from Non-Linear Measurements

Gogna

Majumdar

2016

2016 Data Compression Conference (DCC)

Self Cite

View full text Add to dashboard Cite

This work addresses the problem of estimating proton density and T1 maps from two partially sampled K-space scans such that the total acquisition time remains approximately the same as a single scan. Existing multiparametric non-linear curve fitting techniques require a large number (8 or more) of echoes to estimate the mapsresulting in prolonged (clinically infeasible) acquisition times. Our simulation results show that our method yields very accurate and robust results from only two partially sampled scans (total scan time being the same as a single echo MRI). We model PD and T1 maps to be sparse in some transform domain. The PD map is recovered via standard Compressed Sensing based recovery technique. Estimating the T1 map requires solving an analysis prior sparse recovery problem from non-linear measurements, since the relationship between T1 values and intensity values / K-space samples is not linear. For the first time in this work, we propose an algorithm for analysis prior sparse recovery for non-linear measurements. We have compared our approach with the only existing technique based on matrix factorization from non-linear measurements; our method yields considerably superior results.

show abstract

“…This is an analysis prior sparse recovery problem from non-linear measurements. There is no algorithm to solve (9). In the next section we derive a simple algorithm for the same.…”

Section: T1 Map Estimationmentioning

confidence: 99%

Section: Experiments On T1 Map Estimationmentioning

confidence: 99%

Section: Experiments On T1 Map Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fast Acquisition for Quantitative MRI Maps: Sparse Recovery from Non-Linear Measurements

Gogna

Majumdar

2016

2016 Data Compression Conference (DCC)

Self Cite

View full text Add to dashboard Cite

show abstract

A new method to learn growth curves of beef cattle using a factorization approach

Alonso

Díez

Luaces

et al. 2018

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

The evolution of cattle weight is a very important issue for beef cattle breeders. The weights of the animals of a herd are usually available at different ages and it is intended to predict the trajectory that will follow the weight of each animal. In this paper, we address this problem as a Recommender System. In this case, the users would be the animals, and the items would be the ages of weight measurements. The values of the items would be the measured weights at a given age. As in Recommender Systems the aim is to complete the valuation matrix (weights) in an individualized way (that is, adapted to each animal). A matrix factorization system is devised to learn weights using all the available characteristics of the animals. The weights thus obtained are compared with a linear regression that adequately estimates the general evolution of the herd, but not the individual evolution of each animal. To illustrate the benefits of this approach, we used a real world dataset of cattle of the breed Avileña-Negra Ibérica and crossbreeding with sires of Charolais and Limousin.

show abstract

Matrix factorization from non-linear projections: application in estimating T2 maps from few echoes

Cited by 2 publications

References 19 publications

Fast Acquisition for Quantitative MRI Maps: Sparse Recovery from Non-Linear Measurements

Fast Acquisition for Quantitative MRI Maps: Sparse Recovery from Non-Linear Measurements

A new method to learn growth curves of beef cattle using a factorization approach

Contact Info

Product

Resources

About