Skip to main content

Table 1 Characteristics of patients in the training and test datasets

From: Can computed tomography-based radiomics potentially discriminate between anterior mediastinal cysts and type B1 and B2 thymomas?

Characteristics Training dataset (N = 130) Test dataset (N = 58)
AMC (N = 74) Thymoma (N = 56) P AMC (N = 32) Thymoma (N = 26) P
Age (years) 54.19 ± 8.05 50.71 ± 9.17 0.110 55.81 ± 9.52 44.62 ± 13.04 0.013
Gender    0.213    0.577
 Male 34(45.9%) 18(32.1%)   20 (62.5%) 12 (46.2%)  
 Female 40 (54.1%) 38 (67.9%)   12(37.5%) 14 (53.8%)  
Primary site    0.301    0.041
 Left 44 (59.5%) 30 (53.6%)   22 (68.7%) 18 (69.2%)  
 Right 30 (40.5%) 26 (46.4%)   10 (31.3%) 8 (30.8%)  
Lesion size (cm) 2.72 ± 1.40 3.41 ± 1.26 0.043 2.60 ± 1.34 3.52 ± 1.24 0.069
Unenhanced CT value 33.14 ± 17.59 43.34 ± 9.76 0.004 27.90 ± 17.14 38.06 ± 9.21 0.053
Enhanced CT value 35.26 ± 18.19 67.74 ± 16.34 0.000 34.40 ± 20.04 57.07 ± 16.61 0.003
Change of CT value 2.13 ± 12.45 24.40 ± 12.67 0.000 6.50 ± 14.68 19.01 ± 13.25 0.025
Radscore − 2.52 ± 2.22 2.30 ± 2.72 0.000 − 2.41 ± 1.77 1.96 ± 2.66 0.000
  1. Bold indicates p values below 0.05 were considered significant in both training and test datasets
  2. Continuous variables: represented as mean ± standard deviation (SD); Chi-square test or Fisher’s exact test [number and percentage (%)]
  3. CT computed tomography, HU Hounsfield unit, change of CT value enhanced CT value − unenhanced CT value, AMC anterior mediastinal cysts