PDF(Pubmed)
Abstract：
BACKGROUND: The aim of this study is to explore the utility of cardiac magnetic resonance (CMR) imaging of radiomic features to distinguish active and inactive cardiac sarcoidosis (CS).
METHODS: Subjects were classified into active cardiac sarcoidosis (CSactive) and inactive cardiac sarcoidosis (CSinactive) based on PET-CMR imaging. CSactive was classified as featuring patchy [18F]fluorodeoxyglucose ([18F]FDG) uptake on PET and presence of late gadolinium enhancement (LGE) on CMR, while CSinactive was classified as featuring no [18F]FDG uptake in the presence of LGE on CMR. Among those screened, thirty CSactive and thirty-one CSinactive patients met these criteria. A total of 94 radiomic features were subsequently extracted using PyRadiomics. The values of individual features were compared between CSactive and CSinactive using the Mann-Whitney U test. Subsequently, machine learning (ML) approaches were tested. ML was applied to two sub-sets of radiomic features (signatures A and B) that were selected by logistic regression and PCA, respectively.
RESULTS: Univariate analysis of individual features showed no significant differences. Of all features, gray level co-occurrence matrix (GLCM) joint entropy had a good area under the curve (AUC) and accuracy with the smallest confidence interval, suggesting it may be a good target for further investigation. Some ML classifiers achieved reasonable discrimination between CSactive and CSinactive patients. With signature A, support vector machine and k-neighbors showed good performance with AUC (0.77 and 0.73) and accuracy (0.67 and 0.72), respectively. With signature B, decision tree demonstrated AUC and accuracy around 0.7; Conclusion: CMR radiomic analysis in CS provides promising results to distinguish patients with active and inactive disease.
METHODS: Subjects were classified into active cardiac sarcoidosis (CSactive) and inactive cardiac sarcoidosis (CSinactive) based on PET-CMR imaging. CSactive was classified as featuring patchy [18F]fluorodeoxyglucose ([18F]FDG) uptake on PET and presence of late gadolinium enhancement (LGE) on CMR, while CSinactive was classified as featuring no [18F]FDG uptake in the presence of LGE on CMR. Among those screened, thirty CSactive and thirty-one CSinactive patients met these criteria. A total of 94 radiomic features were subsequently extracted using PyRadiomics. The values of individual features were compared between CSactive and CSinactive using the Mann-Whitney U test. Subsequently, machine learning (ML) approaches were tested. ML was applied to two sub-sets of radiomic features (signatures A and B) that were selected by logistic regression and PCA, respectively.
RESULTS: Univariate analysis of individual features showed no significant differences. Of all features, gray level co-occurrence matrix (GLCM) joint entropy had a good area under the curve (AUC) and accuracy with the smallest confidence interval, suggesting it may be a good target for further investigation. Some ML classifiers achieved reasonable discrimination between CSactive and CSinactive patients. With signature A, support vector machine and k-neighbors showed good performance with AUC (0.77 and 0.73) and accuracy (0.67 and 0.72), respectively. With signature B, decision tree demonstrated AUC and accuracy around 0.7; Conclusion: CMR radiomic analysis in CS provides promising results to distinguish patients with active and inactive disease.
摘要：
背景：本研究的目的是探索心脏磁共振（CMR）影像影像在区分活动性和非活动性心脏结节病（CS）方面的实用性。
方法：根据PET-CMR成像将受试者分为活动性心脏结节病（CSactive）和非活动性心脏结节病（CSactive）。CSactive被分类为具有在PET上的斑片状[18F]氟脱氧葡萄糖（[18F]FDG）摄取和在CMR上存在晚期钆增强（LGE），而CSinactive被分类为在CMR上存在LGE时没有[18F]FDG摄取。在那些被筛选的人中，30例CSactive和31例非CSactive患者符合这些标准.随后使用PyRadiomics提取了总共94个放射学特征。使用Mann-WhitneyU检验在CSactive和CSinactive之间比较了单个特征的值。随后，机器学习(ML)方法进行了测试。将ML应用于通过逻辑回归和PCA选择的两个放射学特征子集（签名A和B），分别。
结果：个体特征的单因素分析没有显着差异。在所有功能中，灰度共生矩阵(GLCM)联合熵具有良好的曲线下面积(AUC)和准确度,置信区间最小,这表明它可能是进一步调查的好目标。一些ML分类器在CSactive和CSinactive患者之间实现了合理的区分。带有签名A，支持向量机和k-邻居表现出良好的性能，AUC（0.77和0.73）和精度（0.67和0.72），分别。带有签名B，决策树显示AUC和准确度约为0.7;结论:CS中的CMR影像组学分析提供了区分活动性和非活动性疾病患者的有希望的结果.
方法：根据PET-CMR成像将受试者分为活动性心脏结节病（CSactive）和非活动性心脏结节病（CSactive）。CSactive被分类为具有在PET上的斑片状[18F]氟脱氧葡萄糖（[18F]FDG）摄取和在CMR上存在晚期钆增强（LGE），而CSinactive被分类为在CMR上存在LGE时没有[18F]FDG摄取。在那些被筛选的人中，30例CSactive和31例非CSactive患者符合这些标准.随后使用PyRadiomics提取了总共94个放射学特征。使用Mann-WhitneyU检验在CSactive和CSinactive之间比较了单个特征的值。随后，机器学习(ML)方法进行了测试。将ML应用于通过逻辑回归和PCA选择的两个放射学特征子集（签名A和B），分别。
结果：个体特征的单因素分析没有显着差异。在所有功能中，灰度共生矩阵(GLCM)联合熵具有良好的曲线下面积(AUC)和准确度,置信区间最小,这表明它可能是进一步调查的好目标。一些ML分类器在CSactive和CSinactive患者之间实现了合理的区分。带有签名A，支持向量机和k-邻居表现出良好的性能，AUC（0.77和0.73）和精度（0.67和0.72），分别。带有签名B，决策树显示AUC和准确度约为0.7;结论:CS中的CMR影像组学分析提供了区分活动性和非活动性疾病患者的有希望的结果.
