Automated approach for fetal and maternal health management using light gradient boosting model with SHAP explainable AI
2024

Automated Fetal and Maternal Health Management Using AI

Sample size: 2126 publication Evidence: high

Author Information

Author(s): Innab Nisreen, Alsubai Shtwai, Alabdulqader Ebtisam Abdullah, Alarfaj Aisha Ahmed, Umer Muhammad, Trelova Silvia, Ashraf Imran

Hypothesis

Can a light gradient boosting model effectively classify fetal health using cardiotocography data?

Conclusion

The proposed model achieved high accuracy and performance metrics, demonstrating its effectiveness in classifying fetal health.

Supporting Evidence

  • The model achieved 99.89% accuracy and 99.88% AUC on the test dataset.
  • SMOTE was used to address class imbalance in the dataset.
  • The proposed model outperformed several other machine learning models in fetal health classification.

Takeaway

This study created a smart computer program that helps doctors check if babies are healthy before they are born by looking at heart rate data.

Methodology

The study used a light gradient boosting machine and synthetic minority oversampling technique (SMOTE) to classify fetal health based on cardiotocography data.

Potential Biases

Potential biases may arise from the dataset not being representative of diverse populations.

Limitations

The study may have biases due to the dataset's class imbalance and the reliance on specific machine learning models.

Participant Demographics

The dataset includes records from expecting mothers with gestational ages between 29 and 42 weeks.

Digital Object Identifier (DOI)

10.3389/fpubh.2024.1462693

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication