| Abstract: | A semi-autonomous computer-aided diagnosis system was developed to characterize the distinctions between malignant tumor types which are known to have general geometric and histological differences: stellate carcinomas and fibroadenomas. The goal was to construct a user-friendly, modular analysis system capable of using only the information contained in the image to distinguish between the two tumor classes. By analyzing tumor circularity, surface roughness, and boundary roughness, the system was successful in achieving distinct class separation of the sample set of 13 fibroadenomas and 13 stellate carcinomas. |
|---|