Rule-Based Labeling of CT Head Image
AIME '97 Proceedings of the 6th Conference on Artificial Intelligence in Medicine in Europe
Modified fuzzy c-mean in medical image segmentation
ICASSP '99 Proceedings of the Acoustics, Speech, and Signal Processing, 1999. on 1999 IEEE International Conference - Volume 06
Distributed Markovian segmentation: Application to MR brain scans
Pattern Recognition
Segmentation of CT Head Images
BMEI '08 Proceedings of the 2008 International Conference on BioMedical Engineering and Informatics - Volume 02
Liver Segmentation from CT Scans: A Survey
WILF '07 Proceedings of the 7th international workshop on Fuzzy Logic and Applications: Applications of Fuzzy Sets Theory
| Hi-index | 0.00 |
In this paper, automatic segmentation and retrieval of medical images are presented. For the segmentation, different unsupervised clustering techniques are employed to partition the Computed Tomography (CT) brain images into three regions, which are the abnormalities, cerebrospinal fluids (CSF) and brain matters. The novel segmentation method proposed is a dual level segmentation approach. The first level segmentation, which purpose is to acquire abnormal regions, uses the combination of fuzzy c-means (FCM) and k-means clustering. The second level segmentation performs either the expectation-maximization (EM) technique or the modified FCM with population-diameter independent (PDI) to segment the remaining intracranial area into CSF and brain matters. The system automatically determines which algorithm to be utilized in order to produce optimum results. The retrieval of the medical images is based on keywords such as "no abnormal region", "abnormal region(s) adjacent to the skull" and "abnormal region(s) not adjacent to the skull". Medical data from collaborating hospital are experimented and promising results are observed.