Abstract #2410

Automatic Segmentation of Laryngeal Cartilages Using Support Vector Machines

R. Reeve Ingle¹, Berhane H. Azage¹, Jolle K. Barral¹, Kie Tae Kwon¹, Edward G. Damrose², Nancy J. Fischbein,^2,3, Dwight G. Nishimura¹

¹Electrical Engineering, Stanford University, Stanford, CA, United States; ²Otolaryngology, Stanford University, Stanford, CA, United States; ³Radiology, Stanford University, Stanford, CA, United States

MR is critical in the staging of laryngeal cancer. However, the presence and extent of cartilage invasion is difficult to assess. In this work, automatic intensity correction is integrated in a support vector machine algorithm, which is used to segment the cartilages from high-resolution MR images of the larynx.