Advanced Neural Network Approach for Fetal Heart Rate Classification with Multi-Attribute Analysis

Electronic fetal monitoring is an essential diagnostic aid in pregnant women for the evaluation of fetal status and possible pathology. In this research, a sophisticated neural network model is suggested as a means for the arrangement of fetal heart rate indicators with multi attribute analysis to assist in maximizing the classification. We thus use an ANFIS that incorporates the capability of learning from the neural network on the one hand and the reasoning of fuzzy logic on the other. The system employs ECG data and the signals are processed by Independent Component Analysis method to extract the fetal signals. Further, PSD transforming of a signal and RR-interval measurements are incorporated to obtain significant signal characteristics making the discrimination more sensitive and specific. The investigational outcomes also show that the proposed model achieves a much higher overall accuracy than the traditional methods, providing an effective means for the fetal heart monitoring in practice.