Objective assessment of pain scale is essential to the understanding of its mechanism as well as the development of pain treatments. The purpose of this work is to develop an objective metrics for measuring endogenous pain scale by analyzing resting magnetoencephalographic data. The data used in this study were collected from fourteen PDM patients at the menstrual phase. Features were extracted by using spectral analysis, temporal complexity analysis and hemispheric asymmetric calculation. Eight features were selected and linear regression was adopted to predict the pain level. The averaged residual error was 0.21 with pain scales ranging from 0 to 10, suggesting that the proposed method could be a reliable indicator for the assessment of endogenous pain scale.