Risk factors for Apgar score using artificial neural networks.

Area of Science:

• Perinatal Medicine
• Computational Biology
• Artificial Intelligence in Healthcare

Background:

• Artificial Neural Networks (ANNs) are established tools for identifying medical risk factors.
• Apgar scores are critical indicators of newborn health immediately after birth.
• Predictive modeling for Apgar scores can aid in early intervention strategies.

Purpose of the Study:

• To introduce and apply Artificial Neural Networks (ANNs) for predicting low Apgar scores.
• To identify significant risk factors associated with low Apgar scores using ANNs.
• To determine the minimal set of influential variables for accurate Apgar score prediction.

Main Methods:

• Utilized a feed forward back propagation Artificial Neural Network (ANN) model.
• Employed the perinatal database from the Perinatal Partnership Program of Eastern and Southeastern Ontario (PPPESO).
• Tested the ANN's ability to generate a strong predictive model and identify influential variables.

Main Results:

• Successfully developed a predictive model for Apgar scores using ANNs.
• Identified key risk factors contributing to low Apgar scores.
• Determined minimal variable sets that maintain ANN performance for Apgar prediction.

Conclusions:

• Artificial Neural Networks (ANNs) demonstrate significant potential for Apgar score prediction.
• The study successfully identified crucial risk factors for low Apgar scores.
• Minimal variable sets can effectively predict Apgar scores without compromising predictive accuracy.