This document presents a study on using a filtered-X least mean square (FXLMS) algorithm to remove various types of noise from electrocardiogram (ECG) signals. The FXLMS algorithm is an adaptive noise cancellation technique that is shown to outperform a standard least mean square (LMS) algorithm in terms of signal-to-noise ratio when removing noise such as baseline wander, powerline interference, muscle artifacts, and motion artifacts from real ECG signals based on simulations using a publicly available ECG database. The key aspects of the FXLMS algorithm and its application to adaptive noise cancelation in ECG signals are discussed.