静息态功能磁共振在轻微型肝性脑病机制及智能诊断中的研究进展OA

Minimal hepatic encephalopathy(MHE),as the early occult stage of hepatic encephalopathy,often presents with unremarkable clinical manifestations and routine blood biochemical parameters.Its diagnosis primarily relies on neurophysiological or neuropsychological tests.Without timely intervention,MHE can progress to overt hepatic encephalopathy,significantly impairing patients'quality of life.In recent years,resting-state functional magnetic resonance imaging(rs-fMRI),with its non-invasive and precise advantages,has provided crucial neuroimaging tools for unraveling the neural mechanisms of MHE.However,existing reviews have predominantly focus on the conventional applications of rs-fMRI in MHE,lacking a systematic synthesis of multi-dimensional brain functional metrics and their integration with artificial intelligence techniques.This review systematically summarizes the latest advancements in utilizing rs-fMRI-based brain functional metrics to investigate the mechanisms of cognitive impairment in MHE.It first elucidates the neuropathological basis of MHE from the perspective of functional connectivity abnormalities,particularly within networks such as the default mode network and the executive control network.Building on this foundation,it further synthesizes the progress in model construction and validation for intelligent MHE diagnosis using machine learning and deep learning methods(e.g.,support vector machines and graph neural networks).Furthermore,it explores the emerging trend of integrating multimodal imaging with artificial intelligence,addressing the gaps present in previous reviews.This article aims to provide a systematic imaging basis for elucidating the neural mechanisms of MHE and to offer new research directions for achieving quantitative and precise diagnosis of MHE.Future research endeavors should focus on multi-center,large-sample validation,multimodal data fusion,and in-depth exploration of pathways for clinical translation.