As an effective method for integrating the information in different forms of medical images, medical image fusion has been commonly used in various clinical applications, such as disease diagnosis and treatment planning. However, the existing medical image fusion methods do not effectively solve the problem of blurred boundaries between different organs, making the fused images more difficult to understand. Therefore, to solve this problem, this paper proposes a medical image fusion model based on multi-scale feature learning and edge enhancement. First, the receptive field is expanded using multiple dilated convolutions with different dilate rates to enable the model to learn more discriminative multi-scale features of the source images. Then, according to the maximum fusion strategy, the source image features are fused to obtain the fused feature. The convolutional layer is used to reconstruct it to obtain the fused image. Further, the edge enhancement module is introduced to enhance the edge information in the fused image to better solve the problem of blurred boundaries between different organs in medical image fusion. The experimental results show that the results obtained by using the proposed method are superior to comparison methods in terms of subjective visual effects and objective quantitative evaluation.