Introduction Specific Aim Volumetric Datasets 3D Bilateral Filtering EM based Classification Classification and Segmentation Collaborators

The EM-based classification algorithm is developed in CCV, The University of Texas at Austin. Classification has been applied to the datasets to identify prostate, seminal vesicles, rectum, and bladder.

This movie shows the classification result of the before dataset, 5 materials are identified.

This movie shows the classification result of the bilaterally filtered before dataset, 7 materials are identified.

This movie shows the classification result of the after dataset, 5 materials are identified.

This movie shows the classification result of the bilaterally filtered after dataset, 9 materials are identified.