In order to improve the application value of the teaching resources of traditional Chinese medicine pharmacology in practical teaching, a method of integration of teaching resources of traditional Chinese medicine pharmacology based on deep learning algorithm was proposed. The teaching resources of TCM pharmacology were collected by association rule mining. From three aspects of text/data, image and video, complete the preprocessing of course teaching resources. The convolutional neural network in the deep learning algorithm is used to extract the characteristics of teaching resources of TCM pharmacology, and finally, the integration of teaching resources is realized according to the measurement results of similarity between resources. Through the performance test experiment, it is concluded that compared with the traditional integration methods, the resource loss rate of the integration results obtained by the optimization design method is reduced by about 3%, and the redundancy of the integration results is significantly reduced.