The accurate delivery of external beam radiation therapy is often facilitated through the implantation of radio-opaque fiducial markers (seeds). Before the delivery of each treatment fraction, seed positions can be determined via volumetric imaging. By registering these seed locations with the corresponding locations in the previously acquired treatment planning CT, it is possible to adjust the patient position or the treatment plan so that seed displacement is accommodated. We present an automatic algorithm that identifies seeds in both planning and pretreatment images and subsequently determines the geometric transformation between the two sets. The algorithm is applied to the imaging series of 10 prostate cancer patients. Each series is comprised of a single multislice planning CT and several megavoltage conebeam CT images obtained immediately prior to a subsequent treatment session. Seed locations were determined for 164 images to within 1mm with an accuracy of 98 ± 6.3%.