目的：驗證與探討電腦刀呼吸同步追蹤系統，在腫瘤因呼吸而造成週期性移動下，治療位置準確度與治療範圍的變化情形，評估該系統於臨床治療上之貢獻及其相關影響。材料與方法：本研究使用具有呼吸同步追蹤功能的電腦刀立體定位放射手術系統。使用移動假體（Motion phantom）模擬病患治療位置因呼吸而移動情形。此假體配置一個由旋轉軸帶動之平台，可作前後方向運動，平台上可供放置內含驗證片的全系統測試用球型立方體。驗證片使用MD-55膠片，掃描驗證片使用Epson expression 1680穿透式平版掃描器。空間位置準確度驗證之分析軟體為End to End test。臨床病患劑量分佈驗證之分析軟體則為Film QA。空間位置準確度分析分靜態、運動不補償及運動自動補償等三種條件進行，由假體靜止狀態到假體移動同時啟動治療設備作動態自動補償，評估電腦刀在各種情況下執行治療計畫時的空間位置準確度情形。在治療位置運動誤差及動態自動補償誤差的評估方面，設計了兩個臨床狀況實驗：一、呼吸週期中不同位置，二、變化呼吸頻率取定位影像之治療準確度評估。臨床劑量驗證則針對肺部、肝臟與攝護腺三個部位，各選擇一位病患之電腦治療計畫，以MD-55劑量驗證膠片，g數值評估方法（3%, 3mm），比對在腫瘤中心點所在的軸狀切面（axial）及矢狀切面（sagittal）劑量量測結果與電腦治療計畫計算結果之差異。結果：假體靜態條件，平均誤差為0.42 ± 0.06 mm。假體週期運動時，在呼吸週期不同時間點取
像的情況下，運動不校正補償與運動校正補償條件下之平均誤差分別為7.59 ± 5.14 mm與0.35
± 0.10 mm，p值小於0.0001。在變化呼吸頻率的情況下，每分鐘20次與每分鐘15次之平均誤差
差距僅有0.12 mm，p值為0.11。臨床病患劑量驗證，肺部、肝臟與攝護腺三組臨床病患之軸狀切面與矢狀切面之2維劑量驗證，g數值評估方法的通過率皆高於99%。結論：在本研究的分析中，電腦刀呼吸同步追蹤系統之空間位置準確度誤差小於0.5 mm，符合立體定位放射手術治療對於空間位置準確度之要求。在處理治療位置會因呼吸而移動的病患時，執行此系統，將可大幅減少照射的範圍，有效降低正常組織的傷害。在臨床病患劑量驗證實驗中，劑量分佈之準確性亦可充分滿足臨床治療的要求。然而使用本系統的治療時間常須1.5~2.5小時，易使病患因久躺不耐而躁動，影響治療位置的準確性。臨床應用時，應斟酌病患的穩定度適當使用即時影像導引的功能，確保治療的品質。

Purpose : To verify the accuracy of CyberKnife® Respiratory Tracking System (SRTS) in tracking the tumor motion during the respiratory cycle and the influence on treatment field; and to evaluate the clinical application of the system.Materials and Methods : The SRTS was a subsystem of the CyberKnife® stereotactic radiosurgery system for respiratory tracking. The motion phantom was used in the respiratory motion experiments,which can imitate regular breathing pattern. The motion phantom contained a motion table and a ball-cube phantom which can hold verification films. We performed the E2E CyberKnife® dose calibration procedure and used MD-55 verification film, flatbed scanner (Epson Expression 1680) and dose analysis software (Film QA) to evaluate the clinical dosimetry and treatment accuracy. The mechanical accuracy of the SRTS was tested and confirmed by three stages of studies including the static phantom, motion phantom and the motion tracking compensation experiments. We also designed two experiments with phantom moved in different phases and frequency of the respiratory cycle to test for the accuracy of the image guide system on motion compensation. The clinical dosimetry verification studies included the lung, liver and prostate cancer patients. The two dimensional dose distribution (sagittal and axial) measured by the MD-55 film was compared with the treatment planning result by a gamma comparison with pass criteria of 3%/3 mm.
Result : The average positional error for the static stage study was 0.42 ± 0.06 mm. The average targeting error with and without the SRTS tracking compensation in the respiratory cycle was 0.35 ± 0.10 mm and 7.59 ± 5.14 mm, respectively (p < 0.0001). The difference of average targeting error
between two respiratory cycles (15/min and 20/min) was 0.12 mm (p = 0.11). In the verification of the axial and sagittal 2-D dose distribution of the three clinical cases, they were 99% compatible.
Conclusion : The spatial positional error of SRTS was less than 0.5 mm which matched the accuracy requirement of stereotatic radiosurgery. By using the SRTS to track and compensate for the motion of lesions during respiratory cycle, we can significantly minimize the volume of the treatment field and the dose of the surrounding critical structures. The accuracy of dosimetry was also proven by our clinical cases experiments. However, additional positional error will be caused by patient who can not
tolerate to the long treatment hours (1.5 to 2.5 hrs).