Triple-Phase Sequential Fusion Network for Hepatobiliary Phase Liver MRI Synthesis

2026-04-24

AI Summary

This study addresses the inefficiency and motion artifacts associated with prolonged delayed scanning in hepatobiliary phase (HBP) MRI. To overcome these limitations, the authors propose TriPF-Net, a novel deep learning framework that synthesizes HBP images by modeling tissue-specific contrast agent uptake and excretion dynamics using temporal information from T1-weighted, arterial-phase, and venous-phase sequences, while maintaining robustness even when parts of the dynamic series are missing. The method innovatively integrates clinical variables—such as age, sex, bilirubin, and albumin levels—and incorporates an enhanced region-guided encoder, a dynamic feature unification module, and a temporally aware fusion loss to improve physiological consistency. Evaluated on both internal and external datasets, TriPF-Net achieves mean absolute errors of 10.65 and 12.41, PSNR values of 23.27 and 23.11, and SSIM scores of 0.76 and 0.78, significantly outperforming existing approaches.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Motion & Tracking
• Machine Learning: Hardware-aware ML

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Gadoxetate disodium-enhanced MRI is essential for the detection and characterization of hepatocellular carcinoma. However, acquisition of the hepatobiliary phase (HBP) requires a prolonged post-contrast delay, which reduces workflow efficiency and increases the risk of motion artifacts. In this study, we propose a Triple-Phase Sequential Fusion Network (TriPF-Net) to synthesize HBP images by leveraging the sequential information from pre-HBP sequences: while T1-weighted imaging serves as the indispensable baseline, the model adaptively integrates arterial-phase (AP) and venous-phase (VP) features when available. By modeling the tissue-specific contrast uptake and excretion dynamics across these three phases, TriPF-Net ensures robust HBP synthesis even under the stochastic absence of one or both dynamic contrast-enhanced sequences. The framework comprises an Enhanced Region-Guided Encoder and a Dynamic Feature Unification Module, optimized with a Region-Guided Sequential Fusion Loss to maintain physiological consistency. In addition, clinical variables, including age, sex, total bilirubin, and albumin, are incorporated to enhance physiological consistency. Compared with conventional methods, TriPF-Net achieved superior performance on datasets from two centers. On the internal dataset, the model achieved an MAE of 10.65, a PSNR of 23.27, and an SSIM of 0.76. On the external validation dataset, the corresponding values were 12.41, 23.11, and 0.78, respectively. This flexible solution enhances clinical workflow and lesion depiction, potentially eliminating the need for delayed HBP acquisition in HCC imaging.

Problem

• hepatobiliary phase
• liver MRI
• gadoxetate disodium
• motion artifacts
• workflow efficiency

Innovation

• Triple-Phase Sequential Fusion
• Hepatobiliary Phase Synthesis
• Dynamic Contrast Enhancement
• Region-Guided Fusion
• Physiological Consistency