US DE COOKIES
Utilitzem cookies necessaries de sistema pel correcte funcionament de la web i cookies opcionals de Google Analytics per obtenir estadístiques de visita (sense obtenir dades personales). + Info
Acceptar cookies
Tornar als resultats
FI
3.654
2021 Medical Image Analysis
Model-informed machine learning for multi-component T(2) relaxometry
Yu T, Canales-Rodríguez EJ, Pizzolato M, Piredda GF, Hilbert T, Fischi-Gomez E, Weigel M, Barakovic M, Bach Cuadra M, Granziera C, Kober T, Thiran JP

Servei limitat a col·laboradors/res de la xarxa de centres de Germanes Hospitalàries. Rebreu un missatge al vostre correu-e amb un enllaç per a la descàrrega del present article.

Abstract

Recovering the T(2) distribution from multi-echo T(2) magnetic resonance (MR) signals is challenging but has high potential as it provides biomarkers characterizing the tissue micro-structure, such as the myelin water fraction (MWF). In this work, we propose to combine machine learning and aspects of parametric (fitting from the MRI signal using biophysical models) and non-parametric (model-free fitting of the T(2) distribution from the signal) approaches to T(2) relaxometry in brain tissue by using a multi-layer perceptron (MLP) for the distribution reconstruction. For training our network, we construct an extensive synthetic dataset derived from biophysical models in order to constrain the outputs with a priori knowledge of in vivo distributions. The proposed approach, called Model-Informed Machine Learning (MIML), takes as input the MR signal and directly outputs the associated T(2) distribution. We evaluate MIML in comparison to a Gaussian Mixture Fitting (parametric) and Regularized Non-Negative Least Squares algorithms (non-parametric) on synthetic data, an ex vivo scan, and high-resolution scans of healthy subjects and a subject with Multiple Sclerosis. In synthetic data, MIML provides more accurate and noise-robust distributions. In real data, MWF maps derived from MIML exhibit the greatest conformity to anatomical scans, have the highest correlation to a histological map of myelin volume, and the best unambiguous lesion visualization and localization, with superior contrast between lesions and normal appearing tissue. In whole-brain analysis, MIML is 22 to 4980 times faster than the non-parametric and parametric methods, respectively.
Formem part de
HH Província Espanya
Contacteu-nos

Avda. Jordà , 8 - 08035 Barcelona
Telèfon: 93 652 99 99 - Ext 1486
E-mail: fundacio@fidmag.com
Formulari de contacte online 

         

 

Reconeixements a la qualitat i l'excel·lència
Darrera modificació: 08/04/2021
AGAUR
CIBERSAM
Generalitat de Catalunya
ISCIII
Unión Europea
MICINN