PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
September 17, 2025Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition0 citations

Value of Deep Learning-Accelerated T1-w Dixon MRI for Upper Abdominal Imaging

View Full Paper
JFJohannes Beat FingerhutTSTobias ScheefHEHannes Engel

Key Points

  • T1DL improved image quality and reduced examination time from 17 seconds to 13 seconds, enhancing patient comfort.
  • Qualitative and independent quantitative assessments supported the conclusion that DL techniques significantly elevate MRI outcomes.
  • The T1DL sequence specifically benefitted abdominal imaging for organs like the liver, spleen, and pancreas, highlighting its clinical relevance.
  • This research underscores the potential of deep learning to optimize MRI practices, particularly in patients who may struggle with standard procedures.

Abstract

Motivation: MRI is crucial for abdominal imaging due to its soft tissue contrast, but conventional sequences are limited by scan time and resolution. Deep-Learning-(DL)-based reconstruction may improve image quality and speed. Goal(s): Does a DL-accelerated T1-weighted sequence (T1DL) enhance image quality and acquisition time for liver, spleen, and pancreas imaging? Approach: This single-center study of 98 patients undergoing post-contrast 3T MRI with standard and T1DL sequences, with qualitative and independent quantitative image assessments. Results: T1DL shortened the examination time from 17s to 13s and enhanced image quality. Impact: DL-accelerated MRI optimizes scan time and image quality, especially benefiting patients with breath-holding difficulties.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Fingerhut et al. (2025) studied this question.

synapsesocial.com/papers/68d4597b31b076d99fa5ce54https://doi.org/10.58530/2025/0104
Ask AI
Helpful
Bookmark
Share
View Full Paper