Prediction of microvascular invasion in solitary hepatocellular carcinoma ≤ 5 cm based on computed tomography radiomics

Table 1 Clinical and imaging features of patients in the training group and verification group

Feature	Training group (n = 124)	Verification group (n = 61)	Z value/χ2 value	P value
Age (yr), median (quartile)	54 (47; 63)	52 (46; 62)	-0.900	0.368
Gender/cases
Male	102	53	0.349	0.555
Female	22	8
Hepatitis B	104	56	1.575	0.210
Liver cirrhosis	84	43	0.044	0.833
AFP (ng/mL)
≤ 20	73	30	1.188	0.276
> 20	51	31
MVI
Negative	82	40	0.000	1.000
Positive	42	21
Tumor size (mm), median (quartile)	36 (28; 44)	34 (27.5; 41)	-0.746	0.456
TTPVI
Negative	56	25	0.145	0.703
Positive	68	36
RVI
Negative	93	40	1.362	0.243
Positive	31	21
Rad-score, median (quartile)	-0.669 (-0.831; -0.546)	-0.640 (-0.780; -0.494)	-0.917	0.359

AFP: α-fetoprotein; MVI: Microvascular invasion; TTPVI: Two-trait predictor of venous invasion; RVI: Radiogenomic invasion; Rad-score: Radiomics score.

Table 2 Comparison of predictive performance between radiomics tags and image features

	Training group		Verification group
	AUC (95%CI)	P value	AUC (95%CI)	P value
Rad-score	0.724 (0.584-0.863)		0.745 (0.655-0.834)
TTPVI	0.590 (0.500-0.679)		0.522 (0.393-0.651)
RVI	0.545 (0.462-0.628)		0.528 (0.401-0.655)
Rad vs TTPVI		0.018		0.043
Rad vs RVI		0.002		0.048

AUC: Area under the curve; CI: Confidence interval; TTPVI: Two-trait predictor of venous invasion; RVI: Radiogenomic invasion; Rad-score: Radiomics score.