A Feasibility Study of Ultra-Short Echo Time MRI for Positive Contrast Visualization of Prostate Brachytherapy Permanent Seed Implants for Post-Implant Dosimetry

Abstract

Purpose: Ultra-short echo time (UTE) imaging is a magnetic resonance imaging (MRI) technique that uses very short echo times (on the order of microseconds) to measure rapid T2 relaxation. An application of UTE is the visualization of magnetic susceptibility-induced shortening of T2 in tissues adjacent to metal, such as prostate tissue with implanted brachytherapy seeds. This study assessed UTE imaging of prostate brachytherapy seeds on a clinical 3T MRI scanner to provide images for post-implant dosimetry. Methods: A prostate tissue phantom was made of gelatin mixed with Gd and other materials to mimic the prostate peripheral zone’s T1 and T2 relaxation times; this phantom was used to investigate the effect of UTE acquisition parameters on brachytherapy seed visibility. A second phantom was made to model prostate tissue surrounded by muscle tissue; this pelvic phantom was implanted with 85 titanium brachytherapy seeds (STM1251, Bard Medical). Both phantoms were scanned on a 3T GE scanner with a 3D UTE pulse sequence and a fast spin echo (FSE) pulse sequence. The average seed SNR, the CNR between seed and prostate material, and visual characteristics of the seeds were assessed. A seed counting procedure was developed based on the visual seed characteristics, and subsequently used by two physicists to locate seeds in UTE images of the pelvic phantom. Results: On 3D UTE images, the metal seeds caused a bright ring-link artifact in adjacent prostate tissue due to susceptibility-induced T2 shortening. The average seed SNR was 15.99±1.52 for UTE compared to 32.32±22.43 for FSE; CNR between seed and prostate was 6.73±1.85 for UTE vs. 23.76±12.87 for FSE. The ring was larger in diameter than a seed itself; apparent seed diameters were 4.65±0.363 mm for UTE compared to 1.46±0.38 mm for FSE. The 3D spatial ring pattern facilitated differentiation of seeds from needle tracks and seed spacers. The two physicists counted 83 and 86 seeds respectively in the UTE images. Prostate boundaries were less well visualized with UTE compared to FSE. Conclusion: With its ability to visualize brachytherapy seeds, UTE imaging appears to provide an alternative approach to CT for seed identification. Compared to fusion of separately-acquired CT images and T2-weighted MR images (for delineation of prostate boundaries), UTE and T2-weighted MR can be acquired in a single imaging session – a convenience to patients while potentially minimizing inter-modality image registration issues. A study in prostate brachytherapy patients of the quality of post-implant dosimetry with UTE