CT-based peritumoral radiomics signatures to predict early recurrence in hepatocellular carcinoma after curative tumor resection or ablation

Table 1 Patient characteristics

Characteristics	Training Set (N = 109)	Validation Set (N = 47)	P
Gender (Male/ Female)	59/50	29/18	0.25
Age (Mean ± SD)	53.2 ± 12.4	55.4 ± 10.6	0.29
Preoperative AFP (Mean ± SD) (ng/mL)	946.7 ± 50,371.4	7891.4 ± 3530.9	1.00
Cirrhosis (positive/negative)	67/42	28/19	0.82
Hepatitis (positive/negative)	96/13	42/5	0.82
Number of nodules (1/≥2)	87/22	33/14	0.35
Lesion diameter (Mean ± SD) (cm)	4.2 ± 2.9	3.9 ± 3.3	0.57
Treatment method (resection/ablation)	33/76	18/29	0.15
ER (%)	50/109(45.9)	25/47(53.2)	0.51
PT-E positive rate (%)	23/109 (21%)	16/47 (34%)	0.13
T-RO risk score (mean ± SD)	0.46 ± 0.28	0.43 ± 0.36	0.58
PT-RO risk score (mean ± SD)	0.46 ± 0.26	0.44 ± 0.29	0.67

SD standard deviation, AFP alpha-fetoprotein. Hepatitis, Hepatitis B or C; ER early recurrence, PT-E peritumoral enhancement, T-RO tumoral radiomics, PT-RO peritumoral radiomics, T-RO risk score refers to the application of T-RO model to the image of the cases in the training and validation sets, and obtain the risk score of each case (the output is the specific value, 0–1). PT-RO risk score refers to the application of PT-RO model to the image of the cases in the training and validation sets, and obtain the risk score of each case (the output is the specific value, 0–1). P-value reflected the differences between the training and validation cohorts, and P values of less than 0.05 (two-sided) were considered statistically significant

ISSN: 1470-7330