Predicting placenta accreta spectrum and high postpartum hemorrhage risk using radiomics from T2-weighted MRI

Table 3 Results of univariate and multivariate logistic regression analysis of predicting PPH risk

Characteristics	Univariate logistic regression analysis		Multivariate logistic regression analysis
Characteristics	OR (95% CI)	P value	OR (95% CI)	P value
R-score	3.863e + 02 (9.1301–1.6348e + 03)	< 0.001*	5.376e + 02 (7.526- 3.840e + 03)	< 0.001*
Maternal Age	0.958 (0.890–1.031)	0.251	0.928 (0.824–1.045)	0.218
Gestational age at MRI	0.904 (0.828–0.986)	0.023	0.903 (0.785–1.038)	0.15
Gravidities	1.405 (1.177–1.677)	< 0.001*	0.798 (0.371–1.716)	0.564
Parities	2.605 (1.763–3.848)	< 0.001*	1.431 (0.235–8.708)	0.697
Prior cesarean deliveries	2.878 (1.955–4.237)	< 0.001*	2.743 (0.613–12.271)	0.187
Placenta Previa	2.112e + 01 (5.035–8.8609e + 01)	< 0.001*	4.019e + 01 (6.459–2.501e + 02)	< 0.001*
History of abortion	1.164 (0.938–1.444)	0.169	1.874 (0.906–3.879)	0.091
History of other uterine operations	0.699 (0.423–1.155)	0.162	2.065 (1.088–3.921)	0.027*
Assisted reproduction	0.275 (0.069–1.094)	0.067 *	0.289 (0.064–1.297)	0.105

Note: * indicates that the variable has a p-value of less than 0.05 in univariate or multivariate analyses and satisfies the characteristic retention condition. The odds ratio (OR) indicates the degree of influence of an independent variable on the probability of the outcome. An OR > 1 suggests that the factor increases the likelihood of the outcome, while an OR < 1 indicates a reduced likelihood. The 95% confidence interval (CI) reflects the precision of the estimate; if the interval does not include 1, the effect is considered statistically significant (p < 0.05)
Abbreviations: OR, odds ratio; R-score, radiomics model output score

ISSN: 1471-2393