The paper introduces the task requirements of classifying positive and negative clinical findings, proposes a positive and negative discrimination method based on clinical findings and its context information enhanced BERT, expounds the general idea and construction path, and analyzes the experimental results. The results show that the method achieves a final Macro-F1 of 78.1% in the test set of the positive and negatived is crimination task of clinical findings in the 2021China Health Information Processing Conference(CHIP 2021).