作者: Ruilin Li , Martin G. Frasch , Hau-Tieng Wu
关键词:
摘要: There is a need for affordable, widely deployable maternal-fetal ECG monitors to improve maternal and fetal health during pregnancy delivery. Based on the diffusion-based channel selection, here we present mathematical formalism clinical validation of an algorithm capable accurate separation from two signal acquired over abdomen. The proposed first algorithm, best authors' knowledge, focusing analysis based abdominal signal, apply it publicly available databases, PhysioNet non-invasive fECG database (adfecgdb) 2013 PhysioNet/Computing in Cardiology Challenge (CinC2013), validate algorithm. state-of-the-art results are achieved when compared with other algorithms suitable modification. Particularly, $F_1$ score R peak detection achieves $99.3\%$ adfecgdb $87.93\%$ CinC2013, mean absolute error estimated locations $4.53$ msec $6.21$ CinC2013.