Development and Validation of a Deep Learning-enabled Single Breath-hold Abbreviated MRI Protocol for Hepatocellular Carcinoma Diagnosis.
Authors
Affiliations (3)
Affiliations (3)
- Department of Radiology, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Shanghai 200032, China.
- Shanghai Institute of Medical Imaging, Fudan University, Shanghai, China.
- Department of Radiology, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in Southern China, Guangzhou, China.
Abstract
Purpose To develop a deep learning-enabled single breath-hold abbreviated MRI (DL-SBH-aMRI) protocol for hepatocellular carcinoma (HCC) diagnosis. Materials and Methods Patients at high risk of HCC from four institutions (January 2019-January 2025) were prospectively and retrospectively included. All patients underwent conventional complete MRI (cMRI) examinations including precontrast T1-weighted imaging (Pre-T1); postcontrast T1WI in arterial (AP), portal venous (VP), and delayed phases (DP); T2-weighted imaging (T2WI); diffusion-weighted imaging (DWI); and apparent diffusion coefficient (ADC) mapping. Four generative models were trained to synthesize full MRI sequences (T2WI, DWI, ADC, AP, VP, DP) from Pre-T1. The best-performing model was selected to generate synthetic sequences, which, combined with Pre-T1 acquired from MRI scanning, formed the DL-SBH-aMRI protocol. Image quality, perceptual realism, and lesion size measurement accuracy were evaluated for DL-SBHaMRI versus cMRI; diagnostic performance at the patient and lesion levels was assessed using a reference standard based on histopathology and imaging findings. Results A total of 1008 patients were included (mean age, 56.8 years ± 11.8 [SD]; 700 males). The diffusion-based generative model (Li-DiffNet) yielded the highest synthetic image quality and was selected as the backbone of DLSBH-aMRI. DL-SBH-aMRI was noninferior to cMRI in subjective image quality scores (4.07-4.16 vs 4.18-4.19, <i>P</i> < .001). DL-SBH-aMRI demonstrated non-inferior diagnostic performance (all <i>P</i> < .001) for HCC at the patient and lesion levels (sensitivity/specificity = 77.9-88.7%/91.6-93.1%) compared with cMRI (sensitivity/specificity = 84.4-92.5%/94.1-95.2%). Conclusion The DL-SBH-aMRI protocol may enable gadolinium-free, rapid liver MRI while preserving fullsequence diagnostic information for HCC diagnosis. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.