Scholarlink

There are a number of potential causes of pregnancy complications, including the mother’s health problems or factors that can impair the fetal development, which in turn can impact the fetal health. When a CTG is carried out during a high-risk pregnancy, the problems that can arise during the pregnancy can be found quickly. Low oxygen levels increase the risk of fetal discomfort, which can be lethal for the developing baby. In order to help health professionals predict the risk to fetal health, this research suggests using various machine learning (ML) models based on Decision Tree (DT) like Decision Tree (DT), DT (ensemble bagging), and DT (Adaboost). The goal of this study is to persuade those in charge of healthcare to use Machine Learning (ML) techniques in their Internet of Things (IoT) infrastructure. Cardiotocography served as the dataset. Bagging with DT (100 DT) was determined to be the top machine learning DT-based model. The optimal accuracy computed to be 97% and the recall results out to be 98% and 97% for the classes with low risk and the classes with high risk, respectively, despite only achieving 87% recall for medium risk. It outperformed other models.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Evaluating Fetal Health Using Models Based on Machine Learning Decision Trees

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Srushti Gajjar,
  • Jigar Sarda,
  • V. Sanjay,
  • Sawsan D. Mahmood,
  • Ahmed J. Obaid

摘要

There are a number of potential causes of pregnancy complications, including the mother’s health problems or factors that can impair the fetal development, which in turn can impact the fetal health. When a CTG is carried out during a high-risk pregnancy, the problems that can arise during the pregnancy can be found quickly. Low oxygen levels increase the risk of fetal discomfort, which can be lethal for the developing baby. In order to help health professionals predict the risk to fetal health, this research suggests using various machine learning (ML) models based on Decision Tree (DT) like Decision Tree (DT), DT (ensemble bagging), and DT (Adaboost). The goal of this study is to persuade those in charge of healthcare to use Machine Learning (ML) techniques in their Internet of Things (IoT) infrastructure. Cardiotocography served as the dataset. Bagging with DT (100 DT) was determined to be the top machine learning DT-based model. The optimal accuracy computed to be 97% and the recall results out to be 98% and 97% for the classes with low risk and the classes with high risk, respectively, despite only achieving 87% recall for medium risk. It outperformed other models.