CaltechAUTHORS
  A Caltech Library Service

Unsupervised learning of MRI tissue properties using MRI physics models

Varadarajan, Divya and Bouman, Katherine L. and van der Kouwe, Andre and Fischl, Bruce and Dalca, Adrian V. (2021) Unsupervised learning of MRI tissue properties using MRI physics models. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20210729-195003832

[img] PDF - Submitted Version
Creative Commons Attribution Non-commercial No Derivatives.

2MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20210729-195003832

Abstract

In neuroimaging, MRI tissue properties characterize underlying neurobiology, provide quantitative biomarkers for neurological disease detection and analysis, and can be used to synthesize arbitrary MRI contrasts. Estimating tissue properties from a single scan session using a protocol available on all clinical scanners promises to reduce scan time and cost, enable quantitative analysis in routine clinical scans and provide scan-independent biomarkers of disease. However, existing tissue properties estimation methods - most often T₁ relaxation, T∗₂ relaxation, and proton density (PD) - require data from multiple scan sessions and cannot estimate all properties from a single clinically available MRI protocol such as the multiecho MRI scan. In addition, the widespread use of non-standard acquisition parameters across clinical imaging sites require estimation methods that can generalize across varying scanner parameters. However, existing learning methods are acquisition protocol specific and cannot estimate from heterogenous clinical data from different imaging sites. In this work we propose an unsupervised deep-learning strategy that employs MRI physics to estimate all three tissue properties from a single multiecho MRI scan session, and generalizes across varying acquisition parameters. The proposed strategy optimizes accurate synthesis of new MRI contrasts from estimated latent tissue properties, enabling unsupervised training, we also employ random acquisition parameters during training to achieve acquisition generalization. We provide the first demonstration of estimating all tissue properties from a single multiecho scan session. We demonstrate improved accuracy and generalizability for tissue property estimation and MRI synthesis.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
https://arxiv.org/abs/2107.02704arXivDiscussion Paper
ORCID:
AuthorORCID
Bouman, Katherine L.0000-0003-0077-4367
Dalca, Adrian V.0000-0002-8422-0136
Additional Information:Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).
Group:Astronomy Department
Subject Keywords:quantitative MRI, relaxation parameters, proton density, multiecho MRI, parameter estimation
Record Number:CaltechAUTHORS:20210729-195003832
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20210729-195003832
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:110072
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:02 Aug 2021 18:44
Last Modified:02 Aug 2021 18:44

Repository Staff Only: item control page