DEXA'11 Proceedings of the 22nd international conference on Database and expert systems applications - Volume Part II
Hi-index | 0.00 |
The content of data of medical images are quite rich and include a large amount of hidden knowledge and patterns which is difficult for users to discover. In order to solve this problem, this paper firstly clusters the region of interest (ROI) extracted from the medical images and marks these images using these clustered ROI. Then we present Diff_FP-growth algorithm with the guidance of medical domain knowledge on these preprocessed medical image dataset to find those association patterns. Finally, we analysis the algorithm by our experiments and obtain a significant conclusion.