CT影像组学在肺结节诊断中的研究进展

Advances of radiomics research based on computed tomography in pulmonary nodules diagnosis

随着肺部CT检查的普遍应用,偶然发现了大量的肺结节,这为其诊断和治疗带来了很大挑战。肺结节的诊断方法主要依赖于传统的CT特征,如结节大小、边缘特征、形态特征及周围特征等。然而这些传统CT特征数量较少,且判断结果易受医师主观判断的影响。影像组学可以定量提取更多的结节特征,具有更好的重复性,因此可以更有效的预测肺结节的良恶性。本文阐述了影像组学的基本过程,并概述了影像组学特征分析用于肺结节诊断方面的优势及其在肺结节管理方面的临床应用价值,包括其在预测恶性肿瘤及区分其组织学亚型等方面的应用前景。影像组学作为一种低成本、无创的方法有潜力为肺结节的早期诊断和个性化治疗提供可参考依据。

With the widespread application of CT lung screening,a large number of pulmonary nodules have been discovered incidentally,which poses great challenges for diagnosis and treatment.The diagnosis of pulmonary nodules relies mainly on traditional CT features.However,the number of these traditional CT features is small,and the results are susceptible to the subjective judgment of the doctor.Radiomics can quantitatively extract greater numbers of nodule features with much better reproducibility,so it can predict whether the nodules are benign and malignan more effectively.This paper summarizes the basic process of radiomics,and outlines the advantages of radiomics feature analysis in the diagnosis of pulmonary nodules and its clinical value in pulmonary nodule management,including its prospect in predicting malignancy and histological subtyping.As a low cost and non-invasive method,radiomics has the potential to provide reference for early diagnosis and personalized treatment of pulmonary nodules.