Scholarlink

<b>Purpose:</b> <p>To facilitate ease of data compilation across diverse populations for training models to synthesize clinical contrast-weighted images from magnetic resonance fingerprinting.</p> <b>Methods:</b> <p>We leverage a semi-supervised training framework using highly accelerated acquisitions of the target contrasts used as ground truths. We utilize complementary randomized data sampling masks across training subjects and contrasts for homogeneous learning in k-space, together with multi-task learning.</p> <b>Results:</b> <p>Our experiments indicate that the proposed method achieves high-quality synthesis with networks trained on retrospectively and prospectively undersampled data of the contrast-weighted images, enabling undersampling up to 12–16<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\times \)</EquationSource> <EquationSource Format="MATHML"><math> <mo>×</mo> </math></EquationSource> </InlineEquation>.</p> <b>Conclusions:</b> <p>The proposed method enables semi-supervised learning for synthesis from MRF with an end-to-end, ultra-fast training data acquisition protocol that is easier to obtain across a large population in clinical settings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Semi-supervision for clinical contrast-weighted image synthesis from magnetic resonance fingerprinting

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月17日

查看全文
  • Mahmut Yurt,
  • Cagan Alkan,
  • Xiaozhi Cao,
  • Congyu Liao,
  • Zihan Zhou,
  • Tolga Cukur,
  • Ali Syed,
  • John Pauly,
  • Shreyas Vasanawala,
  • Kawin Setsompop

摘要

Purpose:

To facilitate ease of data compilation across diverse populations for training models to synthesize clinical contrast-weighted images from magnetic resonance fingerprinting.

Methods:

We leverage a semi-supervised training framework using highly accelerated acquisitions of the target contrasts used as ground truths. We utilize complementary randomized data sampling masks across training subjects and contrasts for homogeneous learning in k-space, together with multi-task learning.

Results:

Our experiments indicate that the proposed method achieves high-quality synthesis with networks trained on retrospectively and prospectively undersampled data of the contrast-weighted images, enabling undersampling up to 12–16 \(\times \) × .

Conclusions:

The proposed method enables semi-supervised learning for synthesis from MRF with an end-to-end, ultra-fast training data acquisition protocol that is easier to obtain across a large population in clinical settings.