直線加速器立體定位放射手術-在執行技術、影像融合與品質保證的新觀點

New perspectives in Delivering Technology, Imaging Registration and Quality Assurance in Linac-based SRS

張秋雄(Chiou-Shiung Chang)

指導教授 ： 莊克士

國立清華大學/原子科學院/生醫工程與環境科學系/博士(2017年)

摘要

立體定位放射手術是一種成熟且廣泛使用治療顱內病灶的技術，一般來說此技術使病人保留良好的腦功能以維持更好的生活品質。同時直線加速器已變成全世界放射治療領域最為普遍使用的設備，藉著結合立體定位放射手術的使用，迅速成為腦部立體定位放射治療入門最快的選項。由於科技快速改良與精確度的大幅提升，使用直線加速器執行立體定位放射手術已經廣為臨床常規使用。此論文整合三個研究部份，保括最佳治療技術、影像融合方法與品質保證改良，分別提出研究結果。基於研究發現，本論文分別針對此三部份為直線加速器立體定位放射手術的準則提出新觀點。
臨床上有數種執行直線加速器立體定位放射手術的技術，第一部份研究的目標就是定義最佳治療技術，由其是腫瘤緊鄰腦幹的困難案例。研究中17個案例都接受過單次高劑量的直線加速器立體定位放射手術治療且腫瘤都緊鄰腦幹。收集的個案中有5種腫瘤類型，實驗分為兩組，一組腫瘤靠近腦幹小於0.5公分，另一組則大於0.5公分但小於1公分。技術比較分為三種，有動態順型弧形治療、強度調控放射治療與體積弧形強度調控放射治療。三種執行技術都包括數個入射角度或是弧形角度，並且有共面與非共面混合使用。使用劑量量化指數評估治療計畫，其中最值得注意的是梯度指數。研究發現當治療腦幹旁的腫瘤，此時最佳放射治療技術為體積弧形強度調控放射治療。
立體定位放射手術電腦治療計畫中，有電腦斷層與核磁共振影像融合這項常規功能，此功能的目的是輪廓勾劃腫瘤標的與正常組織。問題是變形影像融合是否可以應用在立體定位放射手術，目前還有爭論。相應地，在第二部分的研究，另外收集16個治療計畫都執行過立體定位放射手術治療計畫，並且有定形影像融合的結果，同樣地，這16個病患的腫瘤都靠近腦幹。實驗中電腦斷層與核磁共振影像融合都經過定形與變形兩種運算公式分別進行影像融合。腫瘤標的與正常組織的輪廓勾劃依據兩種不同的影像融合公式分別由放射腫瘤專科醫師與合格劑量師描繪出來。評估指數有volume overlapping (VO), Dice similarity coefficient (DSC)與劑量。公式modified-demons deformable algorithm (VARIAN SmartAdapt)為此子實驗之評估標地，驗證其軟體運算在立體定位放射手術電腦治療計畫中的準確度。最終做出其變形影像融合公式在立體定位放射手術電腦治療計畫中是可信賴的結論。
第三部分也是最後的部分，研究調查立體定位放射手術的品質保證，藉由分析非晶矽電子影像驗證系統的劑量特性與每日劑量輸出一致性分析其品質保證使用的可能性。由其地，非晶矽電子影像驗證系統的劑量預測功能應用在線性、照野因子、劑量率與射源至面板距離因子的測試。同時地，量測其他兩種劑量計，包括EBT3底片與市售多點游離腔的每日劑量輸出量測儀。三種劑量每天都須測量，期間由2014年的七月至2015年的二月，一共測量165點的數據， 其結果呈現市售多點游離腔每日劑量輸出量測儀與非晶矽電子影像驗證系統有高度的一致性。也做出非晶矽電子影像驗證系統可以取代複雜EBT3底片的結論。
此論文成功地整合三項子研究並歸納出顯著的結果，對於立體定位放射手術的最佳治療技術、影像融合方法與品質保證改良提供了相當重要的新證據與新觀點。作者相信此論文對於全世界從事立體定位放射手術的從業人員有深刻影響的貢獻。

關鍵字

立體定位放射手術 ； 影像融合 ； 體積弧形強度調控放射治療 ； 變形影像融合 ； 電子影像驗證系統

並列摘要

Stereotactic radiosurgery (SRS) is a well-established technique in the treatment of intracranial lesions, generally leading to better preservation of brain functions and therefore a better quality of life for the patient. Linear accelerators (linacs), meanwhile, have become ubiquitous pieces of equipment in radiotherapy around the world, and the combined use of SRS and linac technology is entry increasingly popular option in stereotactic brain radiotherapy. Dramatic recent improvements in both treatment techniques and linac accuracy performance have enhanced the reliability of SRS, which is now routinely used. This thesis reports the results of three completed studies for linac-based SRS. Based on these findings, the thesis proposes new clinical guidelines and perspectives for linac-based SRS treatment, registration technique and development of the detector for quality assurance.
There are several available forms of linac-based SRS, and the goal of first study in thesis was therefore to identify which of these techniques is optimal when the target is located adjacent to the brainstem. We collected the records of 17 patients with lesions close to the brainstem who had previously been treated with single-fraction radiosurgery. In all, five different lesion catalogues were collected, and the patients were divided into two groups: one consisting of seven individuals with a target-to-brainstem distance of less than 0.5cm, and the other of 10 patients with a target-to-brainstem distance of between 0.5cm and 1cm. Comparisons were then made among the following three types of linac-based radiosurgery: dynamic conformal arcs (DCA), intensity -modulated radiosurgery (IMRS), and volumetric modulated arc radiotherapy (VMAT). All three techniques included multiple non-coplanar beams or arcs, with or without intensity-modulated delivery. Dosimetric metrics were used to evaluate plan quality, with particular attention paid to gradient indexes. This study found that the optimal SRS technique when the tumor was located adjacent to the brainstem was VMAT.
The regular functions of CT-MRI registration include delineation of targets and organs-at-risk (OARs) in radiosurgery planning. The question of whether deformable image registration (DIR) could be applied to SRS in its place remains a subject of debate. Accordingly, study in second part collected data regarding 16 patients who had undergone single-fraction SRS treatment. All their lesions were located close to the brainstem. CT-MRI image sets were registered by both rigid image registration (RIR) and DIR algorithms. The contours of the OARs were drawn individually on the rigid and deformable CT-MRI image sets by qualified radiation oncologists and dosimetrists, and the evaluation metrics included volume overlapping (VO), Dice similarity coefficient (DSC), and dose. A modified-demons deformable algorithm (VARIAN SmartAdapt) was used for evaluation in this study, which verified that software accuracy and concluded reliability of DIR in SRS.
The third and final study investigated utility used in SRS quality assurance, along with the dosimetric characteristics and daily-output consistency of an amorphous-silicon (aSi) electronic portal imaging device (EPID). Specifically, the portal dose prediction function of an aSi-1000 EPID was used to test that panel for linearity, field size, dose rate and source-to-detector distance (SDD). The percentage variance of daily output also was measured using two alternative types of dosimeter: EBT3 film, and a multiple-channels output-check device. A total of 165 daily output readings were taken using each of the three dosimeters between July 2014 and February 2015, and the results suggested that the quality of daily-output monitoring via this aSi EPID was reasonably consistent with that of the commercial daily-output device, and therefore aSi EPIDs should be considered seriously as a replacement for complex film-based procedures.
Taken as a whole, the results of three studies in this thesis provide important new evidence and new perspectives on image registration, delivery techniques and quality assurance for linac-based SRS treatment. As such, the thesis contributed important implications for doctors, patients, medical technicians and hospital administrators around the world, which are discussed in depth.

並列關鍵字

SRS ； fusion ； VMAT ； deformable ； EPID

參考文獻

Abacioglu U, Ozen Z, Yilmaz M, Arifoglu A, Gunhan B, Kayalilar N, Peker S, Sengoz M, Gurdalli S and Cozzi L 2014 Critical appraisal of RapidArc radiosurgery with flattening filter free photon beams for benign brain lesions in comparison to GammaKnife: a treatment planning study Radiat. Oncol. 9 119

Adamson J and Wu Q 2012 Independent verification of gantry angle for pre-treatment VMAT QA using EPID Phys. Med. Biol. 57 6587-600

Al kattar El balaa Z, Foulquier J N, El balaa H, Orthuon A and Touboul E 2009 Quality assurance of enhanced dynamic wedge using the aS500-II, EPID Cancer. Radiother. 13 731-9

Alaei P, Hui S K, Higgins P D and Gerbi B J 2006 The use of a commercial QA device for daily output check of a helical tomotherapy unit Med. Phys. 33 3680-2

Alexander E, Moriarty T M, Davis R B, Wen P Y, Fine H A, Black P M, Kooy H M and Loeffler J S 1995 Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases Journal of the National Cancer Institute. 87 34-40