Cardiotocography data analysis for foetal health classification using Spatial Bayesian Neural Network Optimized with Dwarf Mongoose Optimizer

Pregnancy complications require early detection, but traditional Cardiotocography (CTG) analysis is labor-intensive and error-prone. This manuscript presents Cardiotocography Data Analysis for Foetal Health Classification using Spatial Bayesian Neural Network Optimized with Dwarf Mongoose Optimizer (CDA-FHC-SBNN-DMO). The process involves collecting CTG data, optimizing feature selection with Humboldt Squid Optimization Algorithm (HSOA) and classification using Spatial Bayesian Neural Network (SBNN) to categorize foetal health. Dwarf Mongoose Optimizer (DMO) is used to optimize SBNN. The CDA-FHC-SBNN-DMO method was implemented in Python, outperforms existing methods, achieving improvements of 20.89%, 31.45%, and 28.32% in accuracy, and significant increases in precision and recall.