Prospective comparison of Image quality and diagnostic performance of focal liver lesions between 0.55T MRI and conventional 1.5T MRI
Motivation: MRI is crucial for diagnosing liver lesions. Improved MRI tools, such as 0.55T MRI with deep learning (DL) reconstruction, are anticipated to be applied and validated. Goal(s): We aims to compare the image quality and diagnostic performance with/without DL at 0.55T and 1.5T MRI for liver lesions. Approach: We prospectively collected MRI data from 12 patients, non-contrast enhancement at 0.55T with/without using DL and at 1.5T without using DL-based reconstruction, using enhanced 1.5T scans as the reference standard. Results: 0.55T-DL images demonstrated Signal-to-Noise Ratio (SNR) and subjective ratings comparable to or better than those of 1.5T-nonDL, with no significant difference in diagnostic performance. Impact: Fast 0.55T-DL MRI can enhance the detection and diagnosis of focal liver lesions in the upper abdomen.
Authors
- Dominik M. Nickel (ORCID: https://orcid.org/0000-0002-0360-7233)
- Xiang Wang (ORCID: https://orcid.org/0000-0002-4903-4960)
- Jianxing Qiu (ORCID: https://orcid.org/0000-0002-8808-8021)
- W Li (ORCID: https://orcid.org/0009-0001-2325-5132)
- Kun Zhou (ORCID: https://orcid.org/0000-0002-1599-8630)
- Xiaoming Liu (ORCID: https://orcid.org/0000-0003-3467-5607)
- Xiaoye Wang (ORCID: https://orcid.org/0000-0003-1851-6884)
- Ke Wang (ORCID: https://orcid.org/0000-0002-6760-279X)
- Jinxia Zhu
Publication Details
- Journal
- Proceedings 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 Exhibition
- Published
- 2025-09-16
- DOI
- https://doi.org/10.58530/2025/5224
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00