Brain pathological image retrieval based on correlation of visual feature and report-assisted text

摘要: In the medical domain, digital images are produced in ever-increasing quantities and used for diagnostics therapy. However, it is not sufficient to capture semantic content of an image difficult provide good retrieval results according predefined categories domain less using knowledge. This paper proposes a approach by focusing on report fusion. Firstly feature vectors color texture extracted analyzing low-level features. Secondly, set disjoint tokens with appearance selected define visual vocabulary. Here we use diagnosis from doctor represent each token domain. Finally features as input vector SVM training. And correlation high-level constructed Visual Medical method has been experiment, which concerns brain CT images. result demonstrates improvement, effectiveness, efficiency achieved proposed framework.