Evaluation of an automated seed loader for seed calibration in prostate brachytherapy

Abstract

Automated seed loaders for permanent prostate implants are now commercially available. Besides improved radiation safety, these systems offer seed assay capability and ease of needle loading, making preplanned as well as intra-operative implant procedures more time-efficient. The Isoloader (Mentor Corp., CA) uses individual I125 seeds (SL-125 ProstaSeed) loaded in up to 199 chambers inside a shielded cartridge. The unit performs seed counting and calibration using a builtin solid-state detector. In order to evaluate the reproducibility and accuracy of the calibration process, two test cartridges were measured with the Isoloader itself and compared with a well-type ionization chamber (HDR-1000Plus, Standard Imaging). The air kerma strength measurements for all seeds using the Isoloader had a standard deviation of about 2.7%. For the eight seeds assayed more intensively using both the Isoloader and well chamber, the standard deviations of the measurements for each seed were in the range of 0.8% to 2.8% and 0.6% to 1.3%, respectively. The variation in the Isoloader calibration is attributed to small detector solid angle and bead geometry within seed capsules (verified by radiographs). The reproducibility of the air kerma strength measured by the Isoloader was comparable to that from the well chamber and was clinically acceptable. Seed strength measured with the Isoloader was on average 1% 2% larger than that measured with the well chamber, indicating that the accuracy of the Isoloader was clinically acceptable.

Figure 1

Model SL‐125 ProstaSeed encapsulation design. ⁽⁶⁾ The five silver beads are coated with radioactive ${\mathrm{I}}^{125}$ on the surface.

Figure 2

Left: Air kerma strength of individual seeds in two test IsoCartridges. All seeds were measured multiple times using the Isoloader, as noted in the plot. For each seed, the solid square (■) is the average of the readings, and the error bar represents the range of minimal to maximal reading. The thick solid line is the expected seed strength for each batch, and the two dashed lines stand for the $\pm 7.5\%$ variation. Right: histograms of the left plots. The error bars count statistical error, and the solid line is the Gaussian fit, with the mean value and standard deviation indicated.

Figure 3

(a). Histograms derived from Isoloader measurements for IsoCartridge I. Top: Air kerma strength averaged over 10 readings for each of the 90 seeds, that is, the solid squares (■) in Fig. 2. Bin size 0.005 U. Bottom: Percentage differences between 10 readings and their average for each of the 90 seeds. Bin size 1%. Fig. 3 (b) Similar to Fig. 3(a) but for IsoCartridge II. Bin size is 0.015 U for the top plot and 1% for the bottom plot.

Figure 4

(a) Histograms of Isoloader measurements for the first four seeds in IsoCartridge I. Each seed was measured 200 times. The bin size was 0.002 U for seeds 1 and 2 and 0.004 U for seeds 3 and 4. Fig. 4 (b) Histograms of Isoloader measurements for the first four seeds in IsoCartridge II. Each seed was measured 100 times. Bin size 0.004 U.

Figure 5

(a) Well‐type ionization chamber measurements of seed 2 in IsoCartridge I. Top: original measurements. Solid diamonds (♦) connected by a solid line correspond to “seed up” orientation; open squares (□) connected by a dashed line correspond to “seed down” orientation. For each measurement sequence, the marker (♦ or □) is the average of the four readings taken after each insertion, and the error bar represents the range of minimal to maximal reading. Bottom: histograms derived from the top plot. The error bars count statistical error, and the solid and dashed lines are the Gaussian fits for “seed up” and “seed down” orientations, respectively. Fig. 5 (b) is the same as Fig. 5(a) but for seed 77 in IsoCartridge I. Only one Gaussian fit is applied for both “seed up” and “seed down” orientations.

Figure 6

Transmission radiographs of three seeds from IsoCartridge I show the variability of five ${\mathrm{I}}^{125}$‐coated silver beads inside the seed capsule. The radiographs were taken on a simulator (40 mAs, 65 kVp) at different gantry angles; thus, the seeds are at different orientations. The scale in each image represents the outer length of the seed capsule (4.5 mm). The images were enhanced using MATLAB.