Abstract Archives of the RSNA, 2014
INS159
Registration Method for Gadoxetate Disodium-enhanced MR and Radiation Dose Distribution Maps Using an Extracted Liver-region Mask
Scientific Posters
Presented on December 3, 2014
Presented as part of INS-WEA: Informatics Wednesday Poster Discussions
Toru Higaki PhD, Presenter: Nothing to Disclose
Yuko Nakamura MD, Abstract Co-Author: Nothing to Disclose
Fuminari Tatsugami, Abstract Co-Author: Nothing to Disclose
Tomoki Kimura MD,PhD, Abstract Co-Author: Nothing to Disclose
Yasushi Nagata MD, PhD, Abstract Co-Author: Nothing to Disclose
Kazuo Awai MD, Abstract Co-Author: Research Grant, Toshiba Corporation
Research Grant, Hitachi Ltd
Research Grant, Bayer AG
Research Consultant, DAIICHI SANKYO Group
Research Grant, Eisai Co, Ltd
The delivery of radiotherapy to hepatic tumors has been limited by the high radiation sensitivity of the liver parenchyma. While the development of radiation concentration techniques allows addressing liver tumors by radiotherapy, radiation exposure of the normal parenchyma continues to present problems. Radiotherapy-induced damage depends on the radiation dose. Gadoxetate disodium (EOB) is an MR contrast medium that is specifically taken up by hepatocytes and the degree of EOB enhancement is reflective of focal liver function. We developed a method for the assessment on EOB-enhanced MRI (EOB-MRI) scans of focal liver damage induced by radiation. Here we propose a robust and accurate technique for the fusion of EOB-MRI and radiation dose-distribution maps.
Using a 3T MRI scanner (Vantage Titan 3T, Toshiba Medical Systems, Tokyo, Japan) we performed EOB-MRI in 5 patients considered eligible for stereotactic body radiotherapy (SBRT). Planning CT scans were acquired on a multi-detector CT instrument (LightSpeed RT16, GE Healthcare, Wisconsin, USA). Treatment plans were developed with the aid of a radiation therapy planning system (Pinnacle3, Philips Healthcare). We segmented a liver region from the planning CT using a liver region mask extracted by the physicians involved in the planning of SBRT. After segmentation we applied a deformable registration method (3D Slicer, http://slicer.org) for EOB-MRI to align these scans with the planning CT scans. The EOB-MRI scan and the dose map were aligned because the dose maps and the planning CT scans involved the same coordinate system. We compared the accuracy of the registration method with and without liver segmentation to evaluate the performance of our method. The dice similarity coefficient (DSC) was used to evaluate the registration methods.
With the proposed method, the DSC was 0.58±0.19 in the initial position, 0.86±0.06 after rigid registration, and 0.91±0.04 after deformable registration. With conventional registration these values were 0.58±0.19, 0.86±0.05, and 0.86±0.14, respectively.
Desterilization of the extracted liver region mask yielded robust and accurate deformable registration.
By segmenting organs using an organ region masks created at the time of radiation therapy planning, the accuracy of deformable registration can be improved without additional efforts.
Higaki, T,
Nakamura, Y,
Tatsugami, F,
Kimura, T,
Nagata, Y,
Awai, K,
Registration Method for Gadoxetate Disodium-enhanced MR and Radiation Dose Distribution Maps Using an Extracted Liver-region Mask. Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL.
http://archive.rsna.org/2014/14003742.html