Identification of Glioma Pseudoprogression Based on Gabor Dictionary and Sparse Representation Model

This paper aims to find an effective clinical means to separate glioma pseudoprogression from true recurrence. To this end, the sparse representation method was introduced into the field of medical image processing. The key solution is to combine the training samples into a redundant dictionary. With the sparse decomposition algorithm, the test samples were represented by the combination of the sparse linear coefficients of training samples. Then, a suitable classifier was generated for the classification of sparse atoms. Finally, the author carried out a case study and proved that our method can effectively diagnose pseudoprogression in glioma, and enjoys a good prospect of clinical application.