Dosimetric analysis of stereo and orthogonal reconstruction of interstitial implants.

We have evaluated the performance of stereo and orthogonal computerized three-dimensional implant reconstruction algorithms. Sets of orthogonal and stereo radiographs of a simulated interstitial seed implant were taken. Computer reconstructed coordinates of each source were compared with their measured positions in order to assess the geometric accuracy of the two methods. We find that the effect of random measurement errors on reconstructed coordinates from stereo films is 3-5 times that of the orthogonal method. Although relative individual seed positions may be in error by as much as 1 cm, these errors do not significantly change the shape or area of isodose curves surrounding the implant periphery.