Scholarlink

Cardiac arrest in infants is a critical medical emergency that requires early detection for effective treatment. This project focuses on developing a Cardiac Machine Learning Model (CMLM) to predict neonatal cardiac arrest in the Cardiac Intensive Care Unit (CICU) using advanced statistical methods. The model utilizes physiological indicators and employs techniques such as support vector machines and logistic regression for prediction. Imaging modalities like echocardiography and computed tomography complement the diagnostic process. The proposed CMLM demonstrated strong performance, achieving a delta-p value of 0.912, FDR of 0.894, FOR of 0.076, prevalence threshold of 0.859, and CSI of 0.842 in training and comparable metrics in testing. These results suggest the model’s robustness and reliability. By enabling early detection of cardiac arrest episodes, the CMLM has the potential to significantly reduce mortality and morbidity rates in neonates, improving outcomes for critically ill infants in the CICU.

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

A Novel CNN-RNN-Bayesian Hybrid Model for Predicting Cardiac Arrest in Neonatal Intensive Care

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • C. H. Praveen,
  • K. Rajesh Kumar,
  • M. Satyanarayana,
  • Chiranjeevi S. P. Rao Kandula,
  • V. John Bunyan

摘要

Cardiac arrest in infants is a critical medical emergency that requires early detection for effective treatment. This project focuses on developing a Cardiac Machine Learning Model (CMLM) to predict neonatal cardiac arrest in the Cardiac Intensive Care Unit (CICU) using advanced statistical methods. The model utilizes physiological indicators and employs techniques such as support vector machines and logistic regression for prediction. Imaging modalities like echocardiography and computed tomography complement the diagnostic process. The proposed CMLM demonstrated strong performance, achieving a delta-p value of 0.912, FDR of 0.894, FOR of 0.076, prevalence threshold of 0.859, and CSI of 0.842 in training and comparable metrics in testing. These results suggest the model’s robustness and reliability. By enabling early detection of cardiac arrest episodes, the CMLM has the potential to significantly reduce mortality and morbidity rates in neonates, improving outcomes for critically ill infants in the CICU.