Deep learning reconstruction improves detection of focal liver lesions in hepatobiliary phase compared to conventional EOB-MRI
Frontiers in Medicine · 01.05.2026 · 13
DOI: 10.3389/fmed.2026.1797117 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Medicine, online erschienen am 01.05.2026. Verfasst von Xuewen Peng und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Qilu Hospital of Shandong University, Jinan, Shandong, China. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective
In recent years, the application of Deep Learning Reconstruction (DL Recon) technology in MRI has significantly improved the detection rate of small focal liver lesions (FLLs). This study aims to compare the detection performance of DL Recon EOB-MRI with traditional Gd-EOB-DTPA-enhanced liver MRI during the hepatobiliary phase (HBP) for FLL detection.
Methods
This prospective, single-center study included 53 patients who underwent EOB-MRI during the hepatobiliary phase (HBP). For each patient, four types of images were acquired: 3 mm thickness non-DL reconstruction (Standard Non-DL), 3 mm thickness DL reconstruction (Standard DL), 1 mm thickness non-DL reconstruction (HR Non-DL), and 1 mm thickness DL reconstruction (HR-DL). Lesions were categorized based on signal intensity into high-signal and low-signal types, and classified into five size groups: >30 mm, 20–30 mm, 10–20mm, 5–10mm, and <5 mm. Three abdominal radiologists independently assessed the number of FLLs on HBP images. Pairwise comparisons between the four types of images were made using the Wilcoxon signed-rank test and Generalized Estimating Equations to assess differences.
Results
The intraclass correlation coefficient (ICC) between all groups was >0.90, indicating excellent consistency ( P < 0.001). For high-signal lesions, no differences were observed across groups (except for the total number of high-signal lesions: Standard DL vs. Standard Non-DL: P†† = 0.048, all other P > 0.05). For large low-signal lesions (>10 mm), no significant differences were found between groups ( P > 0.05). For small low-signal lesions (5–10 mm), the Standard DL group showed significantly more lesions than the Standard Non-DL group ( P †† = 0.030). For small low-signal lesions (<5 mm), both DL and Non-DL HR techniques detected significantly more lesions than Non-DL alone ( P † ≤ 0.032).
Conclusion
DL-Recon combined with MR techniques demonstrates significant clinical value in enhancing the detection of small low-signal lesions in EOB-MRI hepatobiliary phase, significantly improving the detection rate of FLLs.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Qilu Hospital of Shandong University, Jinan, Shandong, China.
- Division of Bariatric and Metabolic Surgery, Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China.
- GE Healthcare, Beijing, Daxing District, Beijing, China.
- Qilu Medical Imaging Research Institute, Shandong University, Jinan, China.
Zitieren
APA
Peng, X., Cheng, Q., Tian, R., Lang, L., Li, C., Tao, R., Xu, T., Pylypenko, D., Zuo, L., Yu, D., & Lv, W. (2026). Deep learning reconstruction improves detection of focal liver lesions in hepatobiliary phase compared to conventional EOB-MRI. Frontiers in Medicine, 13. https://doi.org/10.3389/fmed.2026.1797117
Vancouver
Peng X, Cheng Q, Tian R, Lang L, Li C, Tao R, et al. Deep learning reconstruction improves detection of focal liver lesions in hepatobiliary phase compared to conventional EOB-MRI. Frontiers in Medicine. 2026;13. doi:10.3389/fmed.2026.1797117
BibTeX
@article{Peng2026,
author = {Peng, Xuewen and Cheng, Qichao and Tian, Runzhe and Lang, Linlin and Li, Cheng and Tao, Ruixin and Xu, Tianyong and Pylypenko, Dmytro and Zuo, Liping and Yu, Dexin and Lv, Weiwei},
title = {Deep learning reconstruction improves detection of focal liver lesions in hepatobiliary phase compared to conventional EOB-MRI},
journal = {Frontiers in Medicine},
year = {2026},
volume = {13},
doi = {10.3389/fmed.2026.1797117}
}
EUVIMED