PPubMed24 Jun 2021

Model-based multi-parameter mapping

Authors not listed by the archive

Abstract

Title: Model-based multi-parameter mapping Authors: Yaël Balbastre, Mikael Brudfors, Michela Azzarito, Christian Lambert, Martina F. Callaghan, John Ashburner Title: Highlights 1. **•** Model-based estimation of quantitative multi-parameter maps. 2. **•** Maximum-likelihood or Maximum a posteriori solutions. 3. **•** Embedded denoising using a joint total-variation prior. 4. **•** Stable second-order solver using a novel approximate Hessian.

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

A model‑based framework with a novel second‑order optimizer and joint total‑variation regularisation improves multi‑parameter MRI mapping accuracy and convergence.

What’s new

i.Model‑based estimation of quantitative multi‑parameter maps
ii.Maximum‑likelihood and maximum‑a‑posteriori fitting options
iii.Embedded denoising via joint total‑variation (JTV) prior
iv.Stable second‑order solver using a novel approximate Hessian

How they did it

The authors extended a Gauss‑Newton optimisation framework to directly fit the SPGR signal equation using log‑parameter representations and added JTV regularisation across contrasts. They derived a bespoke approximate Hessian for a second‑order optimiser and validated the approach on five repeated MPM scans of a healthy subject using leave‑one‑out cross‑validation against ESTATICS‑based baselines.

What they found

SPGR+JTV consistently outperformed all baseline methods in leave‑one‑out cross‑validation
Regularised maps (SPGR+JTV) showed lower RMSE than non‑regularised SPGR‑ML even with fewer echoes

Caveats

✕No analytical proof that monotonic convergence extends to the multivariate case
✕Validation performed on a single‑subject dataset (five repeats) only

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Model-based multi-parameter mapping

The paper, in five minutes

Ask a question about this paper

Similar papers in all four archives

Similar papers in all four archives

The paper, in five minutes