Predictive value of multimodal ultrasound and artificial intelligence in predicting the effect of induction of labor in full term pregnancy

Induction of labor at full term （IOL） is a commonly used clinical intervention in obstetrics for managing high-risk pregnancies. It primarily involves pharmacological or mechanical methods, such as balloon catheters, to artificially initiate labor, aiming to reduce adverse maternal and neonatal outcomes. One of the primary reasons for IOL failure is the immaturity of the cervix. Although traditional Bishop score has been widely used for assessing cervical maturity, its subjective nature limits its ability to provide accurate predictions. In contrast, two-dimensional and three-dimensional ultrasound measurements have demonstrated higher predictive value. Furthermore, the application of machine learning and deep learning techniques in ultrasound image analysis has significantly enhanced the accuracy of predicting cervical maturity and IOL outcomes. With the rapid advancement of ultrasound technology, multimodal ultrasound techniques have become essential tools for evaluating cervical status and predicting the success of IOL. This review systematically summarizes recent progress in the application of multimodal ultrasound and artificial intelligence techniques for predicting delivery outcomes and timing in IOL at full term, providing a more reliable theoretical foundation and practical reference for clinical diagnosis.