Skip to main content

Table 2 Clinical data of the training and validation groups

From: A radiomics model fusing clinical features to predict microsatellite status preoperatively in colorectal cancer liver metastasis

 

AUC

ACC

SPE

F1

OCM

0.811

0.74

0.74

0.63

ORM

0.684

0.80

0.83

0.71

OJM

0.824

0.83

0.83

0.75

ECM

0.760

0.86

0.79

0.82

ERM

0.744

0.86

0.75

0.72

EJM

0.858

0.84

0.88

0.86

  1. Abbreviations: OCM is a logistic regression model using the clinical information from the raw data; ORM is a logistic regression model using the radiomics information from the raw data; OJM is a logistic regression model combining the clinical and radiomics information from the raw data; ECM is a logistic regression model using the clinical information from the augmented data; ERM is a logistic regression model using the radiomics information from the augmented data; EJM is a logistic regression model combining the clinical and radiomics information from the augmented data. EJM is a logistic regression model that combines clinical and radiomic information from the augmented data