Development and Evaluation of Artificial Intelligence Solutions to Improve Fetal Stability

Abstract

This study aims to proactively prevent risk by immediately notifying hospitals and mothers of relevant information by applying artificial intelligence to fetal heart rate and fetal rhythms to improve fetal stability in the event of changes in fetal health status, abnormalities, emergencies, etc.There is a problem that existing apps diagnose fetal abnormalities by simply measuring changes in heart rate without considering changes in fetal heart rate due to fetal movement.In this study, we developed an app that can accurately diagnose fetal abnormalities by using deep learning data on heart rate changes caused by head compression, uteroplacental insufficiency resulting from placental abruption, and umbilical cord pressure.For deep learning artificial intelligence solutions, this research team applied fetal heart rate and fetal abnormalities determination algorithms, fetal emotional stability index, and emotional stability improvement solution algorithms.To verify the artificial intelligence applied with the algorithm, a Python 3.8 environment was built as a Docker container as the server environment for the analysis system, and the Jupyter and tensorflow deep learning frameworks 2.1 were installed on the container environment.As a result, all of the accurate results were calculated in less than 5 seconds.This research team is in the process of collecting additional clinical data related to fetal heart rate at tertiary medical institutions through collaboration with data-driven hospitals.Eventually, the system will be expanded to include fetal stability using artificial intelligence.

Ähnliche Arbeiten