This article can be cited as A. Turnip, Andrian, M. Turnip, A. Dharma, D. Paninsari, T. Nababan and C. N. Ginting, An Application of Modified Filter Algorithm Fetal Electrocardiogram Signals with Various Subjects, International Journal of Artificial Intelligence, vol. 18, no. 1, pp. 207-217, 2020. Copyright©2020 by CESER Publications An Application of Modified Filter Algorithm Fetal Electrocardiogram Signals with Various Subjects Arjon Turnip 1 , Andrian 2 , Mardi Turnip 2 , Abdi Dharma 2 , Debora Paninsari 3 , Tiarnida Nababan 3 and Crismis Novalinda Ginting 3 1 Department of Electrical Engineering, Universitas Padjadjaran, Bandung, Indonesia; Email: jujhin@gmail.com 2 Faculty of Technology and Computer Science, 3 Faculty of Public Health, Universitas Prima Indonesia, Medan, Indonesia; Email: mdnmylenovotab@gmail.com ABSTRACT Electrocardiogram (ECG) is bio-medical signal processing main theme research focus. This is due to the cause of the massive advance in digital computer that makes automation in detection of several disease by using bio-signals data. Monitoring the fetus heart during labor and pregnancy of the maternal is highly recommended. This process can reduce the risk of the fetus being born later with bad sickness. There are several factors that trammel the ECG result and one of it is the noise from electromyographic (EMG) from electrical maternal muscle activities. The noise could come from the abdominal movement or contractions that cause massive noise that often failed the machine detection accuracy. In this study, a modified filter algorithm is offered to remove the electrical noise from the recorded fetal ECG signals. The results show that the noise was reduced significantly, and the signal become smoother. The experiment was conducted using 3-lead ECG recorder. Keywords: Electrocardiogram, Filtering, Extraction, Bio-signal, Annormalities, Accuracy. Mathematics Subject Classification: 53B50, 90C90 Computing Classification System: I.4 1. INTRODUCTION An electrocardiogram (ECG) is a bio-signal method used to obtain certain information about the subject heart condition, the data recorded is often later interpreted to determine what medication to be done or as a way of prevention by detecting early symptoms of heart problems. Practitioners and physicians that study on cardiac monitoring heavily relies on ECG. Monitoring of subject heart usually is conducted to get data from the subject heart muscle activities in a longer time span. It can provide enough data to find out the faced problems and help the diagnosis. The abnormalities detected can be a deciding factor for either the subject decision or the doctor specialist. It can have several unique characteristics, but the cause must be known in certain, is it caused by gene factor or is it really caused by a disease that affect heart or other heart problems. The ECG records heart muscle electrical activity and then display the data in a form of a linear graph on a screen or on paper. This data could then be interpreted by a medical practitioner to determine the condition of the subject’s