Abstract: Electrocardiogram (ECG) recorder is a precision medical equipment, but affected by the injection molding process, its shell is easy to warp and shrink in the injection molding process, which greatly shortens the service life. To solve this problem, a multi-objective optimization method of injection molding process parameters based on particle swarm optimization least squares support vector machine model (PSO-LSSVM) and improved non dominated sorting genetic algorithm (NSGA-Ⅱ) is proposed. Firstly, based on the samples obtained from orthogonal test, the prediction models of warpage and volume shrinkage are established by PSO-LSSVM algorithm, and then combined with NSGA-Ⅱ for global optimization. Through critical comprehensive analysis of the optimized Pareto optimal solution set, the optimal process parameters are finally obtained. At this time, the minimum warpage of plastic parts is 0.438 3mm, and the minimum volume shrinkage is 8.729%, which is 6.98% and 14.92% lower than that before optimization. At the same time, the injection molding test is carried out for the optimal process parameters. Through the measurement, it is found that the molding quality of plastic parts is good and meets the actual production requirements. The research of this paper provides a theoretical basis for further improving the quality defects of injection molded parts.