Meeting Banner
Abstract #2393

High-Throughput Automatic Tumor Detection and Segmentation in Small-Animal MR Imaging of Patient-Derived Tumor Xenografts

Sudipta Roy1 and Kooresh Shoghi1

1Washington University in St. Louis, Saint Louis, MO, United States

Computer aided tumor detection and segmentation of small animal MR images are prone to spurious lesion, false detection, under segmentation, over segmentation, incompatibles of huge number of images for small animal MR imaging. We propose computer aided method using the combination of fast C-means, morphology and single-phase level set to detect and segment tumor lesions from T2 weighted MR images. Proposed method gives over 90% accuracy when applied to homogeneous tumors.

This abstract and the presentation materials are available to members only; a login is required.

Join Here