Nonlinear FHR baseline estimation using empirical mode decomposition and Kohonen neural network

Automated analysis of the fetal heart rate (FHR) curve plays a significant role in computer-aided fetal monitoring. The first and critical step is the estimation of the FHR baseline. Our study proposes a novel non-linear FHR baseline estimation method using empirical mode decomposition (EMD) and clustering method using Kohonen neural network (KNN). To assess the baseline quality, a comparative study was made against a cited linear baseline estimation algorithm and an existing baseline estimation algorithm of a fetal monitor´s platform using randomly selected FHR signals. The results show that the proposed method reaches a higher subjective satisfaction of the clinical experts with closer basal FHR values and more accurate acceleration/deceleration detections.