Retrospective Study
Copyright ©The Author(s) 2025.
World J Radiol. Jun 28, 2025; 17(6): 108247
Published online Jun 28, 2025. doi: 10.4329/wjr.v17.i6.108247
Table 1 Comparison of clinical information and imaging features in patients with dual-phenotype hepatocellular carcinoma and those with intrahepatic cholangiocarcinoma
Variable
Training set
Testing set
P value
t/z or χ2
P value
t/z or χ2
P value
Sex0.7240.3950.0470.8280.592
Age-2.6280.010-0.1970.8450.676
TBIL0.4580.6470.0660.9470.500
ALT0.4090.683-0.1040.9170.322
AST-2.6370.008-0.7170.4740.254
ALB-1.0130.313-0.9130.3660.763
AFP26.780.00016.5440.0000.758
PLT-0.6770.499-0.4050.6850.083
HBV12.5100.000 2.4220.1200.480
Cirrhosis19.2790.000 3.8780.0490.270
MTD0.6520.5150.0380.9700.750
Margin7.9780.005 0.0070.9310.190
Shape0.8620.3534.5610.0330.128
Dense0.5470.4590.4760.4900.089
Cholangiectasis5.8720.0150.1580.6910.592
Artery within tumor0.2500.6170.0110.9180.376
Nonrim APHE8.8010.0032.4220.1190.311
Nonperipheral washout21.5240.0002.3540.1250.705
Halo sign1.7870.1812.1910.1390.612
Capsule2.4330.2961.8860.3890.879
Cystoid variation and necrosis0.3580.5493.3620.0670.793
Peritumoral enhancement8.9470.0039.4770.0020.730
Table 2 Clinical data and imaging characteristics of dual-phenotype hepatocellular carcinoma and intrahepatic cholangiocarcinoma
Variable
b value
OR value
95%CI
P value
Sex-0.0180.9170.954-1.0030.587
Age0.0081.0080.977-1.0410.601
TBIL-0.0020.9980.992-1.0001.351
ALT1.3583.8890.629-24.0490.144
AST0.9362.551.949-3.3352.934
ALB-0.6710.5110.108-2.4130.397
AFP1.0762.9331.225-7.0210.016
PLT0.3881.4740.423-5.1420.542
HBV-0.9880.3720.089-1.5610.177
Cirrhosis0.1031.1091.062-1.1570.000
MTD0.0011.0010.977-1.0240.06
Margin-0.9880.3720.089-1.5610.177
Shape0.9962.7070.830-8.8282.707
Dense-1.4060.2450.042-1.4280.245
Cholangiectasis0.9852.6780.997-7.1910.051
Artery within tumor1.4214.1370.924-18.5030.064
Nonrim APHE0.0811.0840.314-3.7410.898
Nonperipheral washout0.5981.7650.785-3.9650.068
Halo sign-0.1720.730.39-1.490.267
Capsule0.2281.070.29-1.840.098
Cystoid variation and necrosis0.0731.0760.238-4.8640.924
Peritumoral enhancement0.0971.1021.033-1.1760.003
Table 3 Classification performance of different models
Variables
Clinical-sign model
Radiomic model
Combined model
Training setAUC0.9160.9670.954
95%CI0.850-0.9600.916-0.9910.898-0.984
Sensitivity0.8310.9490.864
95%CI0.735-0.9260.893-1.0000.777-0.952
Specificity0.8420.8420.912
95%CI0.747-0.9370.747-0.9370.839-0.986
Accuracy0.8360.8970.888
95%CI0.834-0.8390.895-0.8980.886-0.890
Testing setAUC0.8110.8450.892
95%CI0.677-0.9070.716-0.9310.774-0.962
Sensitivity0.7310.8080.846
95%CI0.560-0.9010.656-0.9590.707-0.985
Specificity0.7200.8400.840
95%CI0.544-0.8960.696-0.9840.696-0.984
Accuracy0.7260.8240.843
95%CI0.718-0.7330.818-0.8290.838-0.846