Objectives <p>To explore the auxiliary role and application of deep learning-based artificial intelligence (AI) in quality control (QC) evaluation of digital pathological sections.</p> Methods <p>A total of 2137 routine hematoxylin and eosin (HE) slides from the Department of Pathology, the First Affiliated Hospital of Army Medical University, collected between January and December 2022, were scanned into digital slides. Based on slide evaluation standards, these digital slides were scored into four grades: A, B, C, and D. ResNet50\ResNet101, EfficientNet-b5, and swin transformer networks were then employed for classification learning. During model training, pretrained weights on the ImageNet dataset were used as initial values, and the QC data were utilized to perform secondary training and optimization of the models. The dataset was partitioned into training (<i>n</i>=1534), validation (<i>n</i>=174), and test (<i>n</i>=429) sets. The test set was reserved for model validation and performance evaluation, remaining independent from the training and hyperparameter tuning process.</p> Results <p>On the test set, a comparison of the four classification models showed that swin transformer achieved optimal performance across all grades except Grade D, for which ResNet50 demonstrated the best results. Overall, swin transformer exhibited the highest comprehensive performance with an accuracy of 87.22%, an area under the curve (AUC) of 0.88, and a prediction speed of 0.6 s per slide.</p> Conclusions <p>This study preliminarily validates that a deep learning-based auxiliary QC scoring system built on swin transformer performs well in terms of accuracy and timeliness for slide QC evaluation. This method can significantly enhance the efficiency of pathology slide QC and plays an important role in the development of a digital pathology department.</p>