The current lithium battery state of health (SOH) estimation technology primarily focuses on non-life-supporting devices such as electric vehicle power batteries. When applied to artificial heart pump lithium batteries, significant operational differences and the inability of simple models to characterize complex electrochemical reactions limit the accuracy and reliability of SOH estimation. To address the inherent contradiction between the computational complexity of higher-order models and the accuracy of SOH assessment for artificial heart pump lithium batteries, a self-optimizing key health factor estimation algorithm is proposed to establish a lithium battery model. Firstly, to tackle the issue of inaccurate impedance measurement due to the nonlinear currentvoltage characteristics of lithium battery electrochemical systems, a quasi-steady-state electrochemical impedance spectroscopy method is designed, and a self-developed EIS testing device is used to obtain multidimensional impedance information under various temperatures, state of charge, and frequencies. Then, the linear relationship between impedance information and battery health status is analyzed, a minimized impedance objective function is established, and an improved particle swarm optimization algorithm is utilized to solve the optimization problem. Finally, hardware-in-the-loop simulation experiments simulate different conditions under the pulsating mode of artificial heart pump lithium batteries and validate the feasibility and effectiveness of the proposed method. Experimental results show that the proposed algorithm has an estimation error of less than 2% for key health factors under different SOC and temperatures; compared to the standard PSO algorithm, the estimation accuracy of the proposed algorithm is increased by 1.88%, meeting the high-precision model establishment requirements and SOH estimation for artificial heart pump lithium batteries.