Abstract: Electrocardiogram (ECG) signals are fundamental in the diagnosis and monitoring of cardiac conditions, but they are often contaminated by various types of noise, such as power line interference, baseline wander, and muscle artifacts. Traditional noise filtering techniques, including the Least Mean Squares (LMS) algorithm, have been widely used to remove steady-state noise. However, LMS filters alone are less effective when dealing with non-stationary noise. This paper proposes a novel Adaptive Hybrid LMS-Kalman Filter (AHLKF) algorithm that combines the strengths of the LMS and Kalman filtering approaches for enhanced real-time ECG signal processing. The AHLKF dynamically adjusts filter parameters to handle both stationary and non-stationary noise with improved performance. This hybrid approach is designed to preserve critical ECG features while minimizing computational complexity, making it suitable for real-time applications in wearable ECG monitors and portable health dev...