An automated intensity-weighted brachytherapy seed localization algorithm

Brachytherapy has proven to be an effective treatment for various forms of cancer, whereby radioactive material is inserted directly into the body to maximize dosage to malignant tumors while preserving healthy tissue. In order to validate the preoperative or intraoperative dosimetric model, a postimplant evaluation procedure is needed to ensure that the locations of the implanted seeds are consistent with the planning stage. Moreover, development of an automated algorithm for seed detection and localization is necessary to expedite the postimplant evaluation process and reduce human error. Most previously reported algorithms have performed binary transforms on images before attempting to localize seeds. Furthermore, traditional approaches based upon three-dimensional seed shape parameterization and matching require high resolution imaging. The authors propose a new computationally efficient algorithm for automatic seed localization for full three-dimensional, low-resolution data sets that directly applies voxel intensity to the estimation of both seed centroid location and angular seed orientation. Computer simulations, phantom studies, and in vivo computed tomography prostate seed imaging results show that the proposed algorithm can produce reliable results even for low-resolution images.