| With the COVID-19 pandemic, it has become necessary to monitor cardiac ac-tivities for heart patients and everyone. However, the traditional way to use non-portable, intrusive, heavy machines to check the electrocardiography (ECG) is not afeasible solution for a large population. As an alternative, some sensors can collectmagnetocardiography (MCG) signals by measuring the magnetic field produced bythe heart’s electrical currents and converting them into ECG signals. The sensor whichmeasures the MCG signals is susceptible, portable, and consumes low power whichcan be an excellent alternative to monitor cardiac activities. But the challenging partof these sensors would be the noise at the low frequencies because the heart alsooscillates at a low frequency. As the relevant signal and noise share the same spectralproperties, standard linear filtering techniques are inefficient. This work proposes aphysical reservoir computing technique using a circuit that can act as a reservoir anda lightweight machine learning (ML) model to train the output of the circuit to reducethe noise and extract the ECG signals out of the MCG ones.Key words: Chaotic Circuits, Electrocardiography (ECG) and Magnetocardiogra-phy (MCG), LMT Circuit, Machine Learning, Reservoir Computing and Ridge Regres-sion. |