Multi-Dimensional Features Recognition in Radiographic Images: A “Collaborative Discovery” Approach

Tao Xu, Irene Cheng, and Mrinal K. Mandal


Texture analysis, Multi-dimensional features, Computer-aided detection (CAD), Chest radiograph (CXR)


Although many state-of-the-art pattern recognition techniques have been proposed in the literature, detecting clinically significant features in medical images still poses challenges because of their low resolution and superimposed anatomical structures which often obscure the target areas. In this work, we investigate the effectiveness of “collaborative discovery” combining multi-dimensional feature vectors to improve the feature detection rate. Statistical analysis of our experimental results on chest radiographs shows the feasibility of our approach.

Important Links:

Go Back