Fetal state health monitoring using novel Enhanced Binary Bat Algorithm

The timely assessment of hypoxic fetuses by cardiotocography is of utmost significance as deficiency of oxygen in fetuses' leads to fetal distress which can further prove to be fatal or cause neurological diseases. This paper puts forward an Enhanced Binary Bat algorithm, the altered form of Binary Bat Algorithm for the multi-classification problem of Cardiotocography. Subset of optimal and relevant features is selected using the optimized and Enhanced BBA algorithm from cardiotocography dataset. Features selected by various evolutionary algorithms EBBA (Enhanced Binary Bat algorithm), qGWO (quantum Grey Wolf Optimization), Genetic algorithm are 11, 15, 12 simultaneously. EBBA efficiently selects most reduced set of features. The proposed EBBA can be used in feature selection and classification of cardiotocography dataset under different fetus state i.e. normal, suspect and pathologic with an accuracy of 96.21% using machine learning classifier i.e. random forest.

Automated summary

This paper proposes an Enhanced Binary Bat algorithm (EBBA) for feature selection and classification of cardiotocography dataset in the multi-classification problem, which can efficiently and accurately assess the hypoxic condition of fetuses, achieving a high classification accuracy.