Development and validation of a multi-modal ultrasomics model to predict response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer

Table 5 DeLong testing between ultrasomics models

Model	Training cohort				Validation cohort
Model	AUC (95%CI)	P-value in comparison to BUS	P-value in comparison to CEUS	P-value in comparison to Combined_1	AUC (95%CI)	P-value in comparison to BUS	P-value in comparison to CEUS	P-value in comparison to Combined_1
BUS	0. 751 (0.640–0.862)	…	0.654	0.078	0.675 (0.481–0.868)	…	0.156	0.025
CEUS	0. 714 (0.598–0.831)	0.654	…	0.024	0.821 (0.660–0.983)	0.156	…	0.905
Combined_1	0.833 (0.740–0.927)	0.078	0.024	…	0.829 (0.673–0.985)	0.025	0.905	…
Combined_2	0.844 (0.757–0.932)	0.141	0.001	0.774	0.893 (0.780-1.000)	0.008	0.107	0.140

BUS: B-mode ultrasound; CEUS: contrast-enhanced ultrasound; Combined_1: the integration of BUS and CEUS; Combined_2: the integration of BUS, CEUS, and deep learning; AUC: area under the curve; 95% CI: 95% confidence intervals. Statistical significance was demonstrated by P < 0.05

ISSN: 1471-2342