Purpose/Significance To explore the application of deep learning in the quality control evaluation of thinprep cytologic test (TCT), and to improve the efficiency and diagnostic accuracy of examination. Method/Process A total of 2 834 cervical cell smear samples from the First Affiliated Hospital of Shandong First Medical University from January 2017 to December 2023 are collected, and a quality control model is developed, including three tasks:detection of squamous epithelial cell quantity, scoring of slice clarity, and determination of positive or negative. A total of 529 samples are collected from January 2024 to December 2024 to evaluate the sample satisfaction and result consistency of the quality control model. Result/Conclusion The quality control model can effectively reduce the workload of pathologists, increase the concordance rate of TCT diagnostic results, and thereby enhance work efficiency and the level of quality control management.