Incorporating Domain Knowledge into Multistrategical Image Classification

Medical image classification is an important part in domain-specific application image mining because there are several technical aspects which make this problem challenging. In this paper, we firstly quantify the domain knowledge about medical image (especially the symmetry), and then incorporate this quantified measurement into classification. We propose a multistrategical image classification method which utilizes various features by integrating two base classifiers. In our method, a base classifier is trained using the examples misclassified by another base classifier. Therefore, both base classifiers can be collaboratively trained. This complementary method gets a more efficient classification.