影像组学和深度学习在预测肝细胞癌微血管侵犯中的研究进展OACSTPCD

Hepatocellular carcinoma(HCC)is a common malignant tumor that poses a huge threat to human health.Microvascular invasion(MVI)is an important cause of postoperative recurrence and metastasis in HCC patients.Currently,the diagnosis of MVI is mainly confirmed through postoperative pathological examination,which,however,is an invasive method.Non-invasive prediction of MVI before surgery is beneficial for guiding individualized treatment and improving prognosis.In recent years,researchers have achieved remarkable results in predicting MVI in HCC using radiomics and Deep learning(DL)methods.The application of these methods has significantly improved the accuracy of predicting MVI in HCC,providing more precise guidance for patients'treatment.This article elaborated on the research achievements in predicting MVI in HCC based on radiomics and DL in recent years,with the expectation of providing a reference for the clinical realization of precise non-invasive prediction methods,facilitating the formulation of individualized treatment plans,improving patients'prognosis,and promoting the development of diagnostic and treatment technologies for liver cancer.