影像科学与光化学2025，Vol.43Issue(6)：103-113,11.DOI:10.7517/issn.1674-0475.2025.06.14
基于胸肌MDCT影像组学列线图预测COPD急性加重倾向的临床研究
A Novel Nomogram Based on Chest Muscle MDCT Radiomics for Predicting Acute Exacerbation Tendency of COPD
摘要
Abstract
Objective:To develop and validate a radiomics nomogram based on pectoralis muscle MDCT(multi-detector computed tomography)for non-invasive prediction of acute exacerbations in patients with chronic obstructive pulmonary disease(COPD).Methods:A total of 210 COPD patients(including 118 prone to acute exacerbation of COPD and 92 relatively stable COPD)were randomly divided into training(n=146)and validation groups(n=64).Radiomics features of the pectoralis muscle were extracted from chest MDCT images by using 3D-slicer software,filtering the features and constructing an radiomics prediction model.The clinical risk factors were screened by using one-way analysis of variance,a radiomics nomogram was constructed by combining multiple logistic regression with radiomics scors.The performance of the nomogram in discrimination,calibration,and clinical utility was evaluated.Results:Finally,12 radiomics features were selected.The radiomics nomogram demonstrated superior discriminative ability in both the training[AUC=0.932(0.891-0.973)]and validation groups[AUC=0.896(0.816-0.975)],significantly outperforming individual clinical model or radiomics features,P<0.05.The nomogram also showed excellent calibration and clinical practicality.Conclusion:The radiomics nomogram based on pectoralis muscle MDCT images is an effective tool for non-invasive prediction of acute exacerbations in COPD patients,providing valuable support for clinical decision-making.关键词
列线图/多层螺旋计算机体层摄影/影像组学/慢性阻塞性肺病Key words
column chart/multi-detector computed tomography/radiomics/chronic obstructive pulmonary disease分类
医药卫生引用本文复制引用
贺谦,朱婷婷,唐伟,张小明..基于胸肌MDCT影像组学列线图预测COPD急性加重倾向的临床研究[J].影像科学与光化学,2025,43(6):103-113,11.基金项目
南充市科学技术局市校合作专项基金(22SXZRKX0006) （22SXZRKX0006）
川北医学院附属医院2021博士启动基金项目(210832). （210832）
