AarXiv2023Cited 4×

SSL-QALAS: Self-Supervised Learning for Rapid Multiparameter Estimation in Quantitative MRI Using 3D-QALAS

Yohan Jun, Jaejin Cho, Xiaoqing Wang, Michael S. Gee, P. Ellen Grant, Berkin Bilgiç, Borjan Gagoski

Abstract

'P. Ellen Grant' 'Berkin Bilgiç' 'Borjan Gagoski'] 1 Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States 2 Department of Radiology, Harvard Medical School, Boston, MA, United States 3 Department of Radiology, Massachusetts General Hospital, Boston, MA, United States 4 Fetal-Neonatal Neuroimaging & Developmental Science Center, Boston Children's Hospital, Boston, MA, United States 5 Harvard/MIT Health Sciences and Technology, Cambridge, MA, United States

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

SSL-QALAS enables rapid, dictionary‑free reconstruction of multiparametric MRI maps from 3D‑QALAS data in seconds.

What’s new

i.Developed a self‑supervised learning framework (SSL‑QALAS) for dictionary‑free multiparametric MRI mapping.
ii.Showed that SSL‑QALAS achieves accuracy comparable to traditional dictionary matching on phantom and in‑vivo data.
iii.Enabled rapid reconstruction of T1, T2, PD, and IE maps within 10 seconds using a pre‑trained model.
iv.Demonstrated fast scan‑specific fine‑tuning of the model in about 15 minutes, confirming generalizability.

How they did it

A neural network was trained in a self‑supervised manner on 3D‑QALAS acquisitions, eliminating the need for external ground‑truth labels or a dictionary. The trained model was evaluated on an ISMRM/NIST phantom and in‑vivo scans, comparing its T1/T2 estimates to reference methods. Additional experiments assessed pre‑trained, scan‑specific, and transfer‑learning variants.

What they found

Reconstruction of all multiparametric maps completed in ~10 seconds with the pre‑trained SSL‑QALAS model.
Scan‑specific fine‑tuning achieved convergence within ~15 minutes.

§ 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

SSL-QALAS: Self-Supervised Learning for Rapid Multiparameter Estimation in Quantitative MRI Using 3D-QALAS

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