---
OA_place: publisher
OA_type: gold
_id: '21949'
abstract:
- lang: eng
  text: "Cardiac T1 mapping provides critical quantitative insights into myocardial
    tissue composition, enabling the assessment of pathologies such as fibrosis, inflammation,
    and edema.\r\nHowever, the inherently dynamic nature of the heart imposes strict
    limits on acquisition\r\ntimes, making high-resolution T1 mapping a persistent
    challenge. Compressed sensing (CS)\r\napproaches have reduced scan durations by
    undersampling k-space and reconstructing images from partial data, and recent
    studies show that jointly optimizing the undersampling\r\npatterns with the reconstruction
    network can substantially improve performance. Still,\r\nmost current T1 mapping
    pipelines rely on static, hand-crafted masks that do not exploit\r\nthe full acceleration
    and accuracy potential. Furthermore, most existing methods do not\r\nlevarage
    the physical T1 decay model in optimization. In this work, we introduce T1-\r\nPILOT:
    an end-to-end method that explicitly incorporates the T1 signal relaxation model\r\ninto
    the sampling–reconstruction framework to guide the learning of non-Cartesian trajectories,
    cross-frame alignment, and T1 decay estimation. Through extensive experiments\r\non
    the CMRxRecon dataset, T1-PILOT significantly outperforms several baseline strategies
    (including learned single-mask and fixed radial or golden-angle sampling schemes),\r\nachieving
    higher T1 map fidelity at greater acceleration factors. In particular, we observe
    consistent gains in PSNR and VIF relative to existing methods, along with marked\r\nimprovements
    in delineating finer myocardial structures. Our results highlight that optimizing
    sampling trajectories in tandem with the physical relaxation model leads to both\r\nenhanced
    quantitative accuracy and reduced acquisition times. Code for reproducing all\r\nexperiments
    and results is available at https://github.com/tamirshor7/T1-PILOT"
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Tamir
  full_name: Shor, Tamir
  last_name: Shor
- first_name: Moti
  full_name: Freiman, Moti
  last_name: Freiman
- first_name: Chaim
  full_name: Baskin, Chaim
  last_name: Baskin
- first_name: Alexander
  full_name: Bronstein, Alexander
  id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
  last_name: Bronstein
  orcid: 0000-0001-9699-8730
citation:
  ama: 'Shor T, Freiman M, Baskin C, Bronstein AM. T1-PILOT: Physics-informed learned
    optimized trajectories for T1 mapping acceleration. In: <i>Medical Imaging with
    Deep Learning</i>. Vol 315. ML Research Press; :1969-1982.'
  apa: 'Shor, T., Freiman, M., Baskin, C., &#38; Bronstein, A. M. (n.d.). T1-PILOT:
    Physics-informed learned optimized trajectories for T1 mapping acceleration. In
    <i>Medical Imaging with Deep Learning</i> (Vol. 315, pp. 1969–1982). Taipei, Taiwan:
    ML Research Press.'
  chicago: 'Shor, Tamir, Moti Freiman, Chaim Baskin, and Alex M. Bronstein. “T1-PILOT:
    Physics-Informed Learned Optimized Trajectories for T1 Mapping Acceleration.”
    In <i>Medical Imaging with Deep Learning</i>, 315:1969–82. ML Research Press,
    n.d.'
  ieee: 'T. Shor, M. Freiman, C. Baskin, and A. M. Bronstein, “T1-PILOT: Physics-informed
    learned optimized trajectories for T1 mapping acceleration,” in <i>Medical Imaging
    with Deep Learning</i>, Taipei, Taiwan, vol. 315, pp. 1969–1982.'
  ista: 'Shor T, Freiman M, Baskin C, Bronstein AM. T1-PILOT: Physics-informed learned
    optimized trajectories for T1 mapping acceleration. Medical Imaging with Deep
    Learning. MIDL: Medical Imaging with Deep Learning, PMLR, vol. 315, 1969–1982.'
  mla: 'Shor, Tamir, et al. “T1-PILOT: Physics-Informed Learned Optimized Trajectories
    for T1 Mapping Acceleration.” <i>Medical Imaging with Deep Learning</i>, vol.
    315, ML Research Press, pp. 1969–82.'
  short: T. Shor, M. Freiman, C. Baskin, A.M. Bronstein, in:, Medical Imaging with
    Deep Learning, ML Research Press, n.d., pp. 1969–1982.
conference:
  end_date: 2026-07-10
  location: Taipei, Taiwan
  name: 'MIDL: Medical Imaging with Deep Learning'
  start_date: 2026-07-08
corr_author: '1'
date_created: 2026-06-07T22:01:36Z
date_published: 2026-03-17T00:00:00Z
date_updated: 2026-06-08T08:05:24Z
day: '17'
ddc:
- '000'
department:
- _id: AlBr
has_accepted_license: '1'
intvolume: '       315'
keyword:
- Cardiac T1 Mapping
- Trajectory Optimization and Reconstruction
- PhysicsInformed Deep-Learning
language:
- iso: eng
license: https://creativecommons.org/licenses/by/4.0/
main_file_link:
- open_access: '1'
  url: https://openreview.net/forum?id=nZaPtHbd6N#discussion
month: '03'
oa: 1
oa_version: Published Version
page: 1969-1982
publication: Medical Imaging with Deep Learning
publication_identifier:
  eissn:
  - 2640-3498
publication_status: accepted
publisher: ML Research Press
quality_controlled: '1'
related_material:
  link:
  - relation: software
    url: https://github.com/tamirshor7/T1-PILOT
scopus_import: '1'
status: public
title: 'T1-PILOT: Physics-informed learned optimized trajectories for T1 mapping acceleration'
tmp:
  image: /images/cc_by.png
  legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
  name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
  short: CC BY (4.0)
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 315
year: '2026'
...