Answers are grounded in this paper and cite their sources.
Early identification of women at risk for persistent postpartum diastasis recti abdominis (DRA) remains challenging because conventional assessment relies mainly on postpartum inter-rectus distance (IRD) and fails to capture early microstructural and biomechanical alterations of the abdominal wall. We aimed to develop and validate an early-pregnancy structure–mechanical model integrating ultrasound radiomics and shear wave elastography (SWE) to predict persistent DRA at 6 months postpartum. In this dual-center study, 390 women from a retrospective cohort were randomly divided into training and testing sets (7:3), and 102 women from an independent prospective cohort. At 5–8 weeks’ gestation, standardized umbilical-level ultrasound of the rectus abdominis–linea alba complex was performed, including B-mode imaging for radiomics analysis and SWE for Young’s modulus measurement. Radiomic scores were derived for the rectus abdominis and linea alba. Persistent DRA, defined as IRD >2 cm at 6 months postpartum, was the primary endpoint. Clinical, radiomics, and structure–mechanical models were built using logistic regression and assessed by discrimination, calibration, and decision curve analysis. Persistent DRA occurred in 40.3% of the training set, 41.0% of the testing set, and 43.1% of the external validation set. Women with persistent DRA had higher radiomic scores and lower Young’s modulus values than those without DRA (all P < 0.001). Radscore1 was positively correlated with postpartum IRD ( r = 0.54), whereas Young’s modulus was negatively correlated with IRD ( r = − 0.68; both P < 0.001). The structure–mechanical model achieved AUCs of 0.893, 0.858, and 0.847 in the training, testing, and independent prospective validation cohort, respectively, outperforming the clinical and radiomics models. Risk-gradient analysis showed a progressive increase in persistent DRA incidence from 18.6% in the low-radscore/high-Young’s modulus group to 71.8% in the high-radiomic-risk/low-stiffness group ( P for trend < 0.001). An early-pregnancy structure–mechanical model combining ultrasound radiomics and SWE of the rectus abdominis–linea alba complex model demonstrated promising discrimination and calibration for predicting persistent DRA, may inform risk-stratified research protocols for future interventional studies.
No takes yet. Share an insight, caveat, or question.
Du et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: