This study uses Heart Rate (HR) and Heart Rate Variability (HRV) as biometric modalities for user authentication. Traditional knowledge-based and possession-based authentication methods suffer from significant limitations, such as vulnerability to security breaches and the inconvenience of physical devices. Biometrics, such as HR and HRV, provide a unique and potentially more secure alternative. By collecting HR and HRV data from 16 participants using a wearable device, this study evaluates the feasibility of using these physiological signals for continuous and transparent authentication. The data was processed using feature extraction techniques, including continuous wavelet transformation and statistical analysis, to identify critical features for classification. A Siamese Neural Network (SNN) is employed to classify the data, achieving significant accuracy improvements when HR data is integrated with HRV. The results demonstrate that HR-based biometric authentication provides higher accuracy than HRV alone, with odds ratios confirming the system's ability to correctly authenticate legitimate users while minimising errors when HR and HRV data is merged.