As a complex anisotropic medium, variation in birefringence within biological tissues is closely associated with numerous physiological behaviors and phenomena. In this Letter, we propose a polarization feature fusion method and corresponding polarimetric parameters, which exhibit excellent performance of capturing the birefringence dynamic variation process in complex anisotropic media. By employing the feature fusion method, we combine and transform polarization basis parameters (PBPs) to derive fused polarization feature parameters (FPPs) with explicit expressions. Subsequently, we conduct Monte Carlo (MC) simulation to demonstrate the effectiveness of the proposed FPPs from two variation dimensions of birefringence direction θ and modulus Δn. Leveraging mathematical modeling and linear transformations, we investigate and abstract their response patterns concerning θ and Δn. Finally, the experiments confirm that the FPPs show superior adaptability and interpretability in characterizing the birefringence dynamic process of turbid media. The findings presented in this study provide new, to the best of our knowledge, methodological insights of information extraction for computational polarimetry in biomedical research.