The development of efficient search engine queries for biomedical images, especially in case of query-mismatch is still defined as an ill-posed problem. Vector-space model is found to be useful for handling the query-mismatch issue. However, vector-space model does not consider the relational details among the keywords and biomedical image search space is not evaluated. Therefore, in this paper, we have proposed a deep learning based fusion vector-space based model. The proposed model enhances the biomedical image query similarity matching approach by fusing the vector space model and convolutional neural networks. Deep learning model is defined by converting the vector-space model to a classification model. Finally, deep learning model is trained to implement the search engine for biomedical images. Extensive experiments reveal that the proposed model achieves significant improvement over the existing models.