Objective The purpose of this study was to compare the adenomyosis models in nude mice generated by four different methods,and to find out an optimal modeling method, and to provide an ideal animal model for exploring pathogenesis and experimental treatment of uterine adenomyosis.Methods 1. 80 female healthy nude mouse were divided randomly into 4 groups:Intraperitoneal implantation group,subcutaneous implantation group, intraperitoneal injection group, and subcutaneous injection group. The transplants were taken for pathological examination at 4 weeks after surgery.Results The success rate of intraperitoneal implantation group was 95%,and that of the subcutaneous implantation group was 45%,while the success rate of intraperitoneal injection group and subcutaneous injection group was 0%.Conclusions Establishment of a nude mouse model of uterine adenomyosis by intraperitoneal implantation method has a high success rate and with good stability, and is an ideal mouse model of human-derived uterine adenomyosis.