Fetal heart rate monitoring through non-invasive electrocardiography is of great relevance in clinical practice to supervise the fetal health during pregnancy. However, the analysis of fetal ECG is considered a challenging problem for biomedical and signal processing communities. This is mainly due to the low signal- to-noise ratio of fetal ECG and the difficulties in cancellation of maternal QRS complexes, motion, etc. This paper presents a survey of different unsupervised classification algorithms for the detection of fetal QRS complexes from abdominal ECG signals. Concretely, clustering algorithms are applied to classify signal features into noise, maternal QRS complexes and fetal QRS complexes. Hierarchical, k-means, k-medoids, fuzzy c-means, and dominant sets were the selected algorithms for this work. A MATLAB GUI has been developed to automatically apply the clustering algorithms and display FHR monitoring. Real abdominal ECG signals have been used for this study, which validate the proposed method and show high efficiency.
