A robotic device for MRI-guided prostate brachytherapy

Abstract

One of the treatment options for prostate cancer is brachytherapy with iodine-125 sources. In prostate brachytherapy a high radiation dose is delivered to the prostate with a steep dose fall off to critical surrounding organs. The implantation of the iodine sources is currently performed under ultrasound guidance, but MRI (Magnetic Resonance Imaging) offers a better image quality that could be employed to further improve source placement precision and prostate dose distribution. Due to the limited amount of space inside a closed bore MRI scanner the currently used manual implantation technique is impossible. Therefore a robotic device, which can be placed between the patient’s legs, is being developed. Prostate motion is one of the main causes of seed misplacement. Because this motion is more or less random there is no practical solution to take it into account during planning of the treatment (chapter 2). To diminish prostate motion a new needle insertion method was developed. Instead of pushing the needle into the prostate, the needle is tapped into the prostate with a high velocity. A tapping device was built and tested on a piece a beef (chapter 3). The tests showed less beef motion with higher needle insertion velocities. After the phantom experiments a clinical study was performed to compare prostate motion during needle insertion for pushing and tapping the needle into the prostate (chapter 4). The mean prostate motion was 5.6~mm when the needle was pushed and 0.9 mm when the needle was tapped into the prostate. We expect that prostate movement will be further reduced when the tapping device is operated at higher momentums. To be able to deliver the iodine seeds at the pre-planned position a good image of the prostate, the needle and the seeds is necessary. The only currently available commercial MRI compatible needles are made of titanium, which still gives a rather large artefact at the tip. This makes it difficult to determine the exact position of the delivered seeds (dimensions 4.5 x 0.8 mm), which is of importance for accurate dose delivery in the prostate. In chapter 5 a simulation study was performed to investigate the influence of different needle materials on the seed artefact. Only with a plastic needle it was possible to distinguish the seed artefact. When the middle of the seed position is taken as the middle of the seed artefact, the seed position can be determined with an accuracy of 0.4 mm on Gradient Echo images. In chapter 6 the first prototype of the MRI compatible robotic device is described. MRI compatibility tests proved the working of the robotic device in a magnetic field without distorting the MR images. In the near future pre-clinical experiments will be performed to test the functionality and the needle and seed placement accuracy of the robotic device. Thanks to the new needle insertion method and the MRI-guidance we think that more accurate seed placement and therefore a better dose distribution is possible than currently achieved.