Cardiovascular disease is one of the leading causes of morbidity and mortality worldwide. Timely and reliable risk assessment is crucial for reducing disease risk and ensuring safety. The aim of this research is to propose an efficient and convenient risk assessment method for cardiovascular disease. In this research, Fourier transform infrared attenuated total reflectance spectra and Raman spectra of 108 whole blood samples were collected for the construction and evaluation of risk assessment models. To address the issue of low efficiency in risk assessment models based on traditional PLS, siPLS, and other feature extraction algorithms, a chemical bond-driven synergy interval partial least squares algorithm (CBDsiPLS) is proposed for feature extraction, and combined with machine learning to construct a risk assessment model using single data sets. The test results show that the proposed method outperforms traditional feature extraction algorithms. In addition, by utilizing the complementary information from mid-infrared and Raman spectroscopy, a risk assessment model for fused data was established through feature-level information fusion combined with machine learning methods. The final fused data risk assessment model achieves an accuracy of more than 90%, a sensitivity of more than 80%, and a specificity of 95%. The experimental results show that the proposed method can effectively assess the risk of cardiovascular disease.