http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
SURFACE TEXTURE ANALYSIS FOR 3D SHAPE ESTIMATION OF CELL NUCLEI IN NUCLEAR GRADING
Hyunju Choi,Taeyun Kim,Uno Tak,Haegil Hwang,Heungkook Choi 한국멀티미디어학회 2007 한국멀티미디어학회 국제학술대회 Vol.2007 No.-
Objective: To find surface texture descriptors for estimating 3D shape of cell nuclei in nuclear grading of renal cell carcinoma and evaluate the feasibility of such quantitative analysis. Methods: After surface rendering, the surfaces of 3D cell nuclei consist of triangle meshes. Our method is based on the relationship between the normal vectors of adjacent triangles. In order to derive the relationship of them, we computed angles between adjacent normal vectors and made an angle feature histogram. Based on this feature histogram, we defined surface texture descriptors and evaluated their significance in nuclear grading. Results and Conclusion: We found a statistically significant correlation between nuclear grade and surface texture descriptors of cell nuclei. It means that surface texture descriptors can be important diagnostic factors to improve the accuracy of grading renal cell carcinoma.