A Study on Benchmarks for Ectopic Pregnancy Classification Using Deep Learning Based on Risk Criteria

Abstract

Of the entire pregnancies, a progressively widespread incidence reported at roughly 1.5–2.0% is called ectopic pregnancy (EP). When a fertilized egg nurtures outside a woman’s uterus, elsewhere in her belly, the EP occurs. It is also called extrauterine pregnancy. In women of reproductive age, it engenders morbidity but rarely mortality. A lively area of study is the expansion of new informatics techniques that are concentrated on ameliorating pregnancy outcomes. The diverse kinds of EPs and the risk factors (RF) related to the EP, ultrasound findings, common diagnostic methods, and the involvement of deep learning (DL) algorithms in EP classification are evinced in this survey. At first, detailing the diagnostic criteria for EP along with non-EP is the goal of this paper. Delineating the utilization of DL methodologies and ascertaining how to employ them for the women to suitable follow-up if they possess a pregnancy of unknown location (PUL) is the next objective. Nevertheless, the studies associated with DL approaches on EPs have not yet been discovered. Here, for the purpose of creating a model to predict and classify EP, diverse kinds of artificial intelligence (AI) algorithms have been examined.