| Lung cancer is one of the most malignant tumors with the highest morbidity and mortality in the whole world.It has become the leading cause of cancer death in both men and women,and nearly 1 million 200 thousand people die from lung cancer each year,the death toll is more than three in breast cancer,colon cancer,and prostate cancer.The latest data statistics shows in 2013,there are about 1 million 700 thousand new cancer cases,and 600 thousand cases of cancer deaths in the United States,male lung cancer accounts for about 28% of them,women accounted for about 26%.In recent years,with the development of radiotherapy technology and the improvement of radiotherapy equipment,three-dimensional conformal radiotherapy(3D-CRT)and intensity modulated radiotherapy(IGRT)are two kinds of new methods.The sharp change in the dose gradient,especially near the edge of the organs,is becoming more and more potential for the risk of organs.So it is important to do the localization more accurately to minimize the error and get the best treatment effect.Image guided radiotherapy(IGRT)technique is adopted to measure setup errors of tumor and organs at risk before and during radiotherapy.Kilo-voltage cone beam CT is a commonly used method of IGRT,image acquisitied is clear enough to match with plan CT scans and then correct setup errors.46 patients of lung cancer were collected in this research.We acquired CBCT scans before IMRT,matched with plan CT scans in different image registration methods,then the setup errors were studied,then we placed system error(?)and random error(?)into Van Herk formula to calculate set-up margins to get further clinical guidance.The contents of 2 sections are summarized as follows.Part 1 A study of comparison of different image registration methods andsetup errors in KV CBCT image-guided radiotherapy for lungcarcinomaObjective: To compare different methods of image registration of KV CBCT scans and planning CT scan and study the setup errors of lung carcinoma radiotherapy in our hospital.Methods:1 46 patients of lung carcinoma were chosen at random.Patients of conventional fraction accepted first CBCT scan before first IMRT,and then one to two times CBCT scan/ week;Patients of hypofraction(divided dose >500cGy)accepted CBCT scan before each treatment.Varian—IX linear accelerator on-board imaging system(OBI)was used to acquire CBCT scans before delivery.378 CBCT scans were collected.2 In the offline study,each CBCT was registered with relative planning CT using automatic and manual methods.In the automatic method,we chose soft tissue matching mode,and entirety was used as a surrogate for image registration,after registration screen automatically displays setup errors around the right-left(X),anterior-posterior(Y),superior-interferior(Z)directions.3 In manual method,we chose GTV as a surrogate for image registration,We do the registration until the CBCT scans and planning CT were coincidenced completely.After registration screen automatically displays setup errors around the X?Y?Z directions.Results:1 In the automatic method,The absolute maximum of all the patients at X,Y,and Z axis direction was 13mm?19mm?8mm.Within 3mm and 5mm of which accounted for 79.1%?54.8%?61.5%and95.5%?73.6%?95.5%respectively.There was only 0.3%?9.5%?0% of setup errors were more than10 mm.2 In the manual method,The absolute maximum of all the patients at X,Y,and Z axis direction was 13mm?21mm?10mm.Within 3mm and 5mm of which accounted for74.3%?54.2%?59.5%and88.9%?70.1%?88.9%respectively.There were only 1.3%?10.8%?0%of errors more than 10 mm.3 In the automatic method,the setup errors were different in Y?Z axis between central type and peripheral type(Y axis P=0.048 Z axis P=0.034).The setup errors were different in Y?Z axis between upper lobe type and lower lobe type(Y axis P=0.044 Z axis P=0.033).4 In the manual method,the setup errors were different in Y? Z axis between central type and peripheral type(Y axis P=0.038 Z axis P=0.026).The setup errors were different in Y?Z axis between upper lobe type and lower lobe type(Y axis P=0.014 Z axis P=0.022)Conclusions: There were setup errors during the treatment process of 46 patients.The most obvious were in the anterior-posterior(Y).Errors of superior-interferior(Z)were the next one,and those of the right-left(X)were the smallest.Both in automatic and manual method.There were significant differences in the Y axis and Z axis between setup errors of upper lobe lung cancer and lower lobe lung cancer.The thoracic diameter,skin relaxation,swinging position during the rotation of the body is due to a difference.We advise doctors to fully consider the incidence of the different parts when individualized discharge.Part 2 A study of set-up margins in KV cone-beam computed tomographyimage-guided radiotherapy for lung carcinomaObjective: To use Van Herk formula to calculate setup margins of lung cancer in the treatment in our hospital by system error(?)and random error(?)got in the first part.Methods:1 The method of acquisition of system error(?)and random error(?)is the same as the first part.According to the regulation of ICRU,we delineated GTV,CTV and organs at risk,then according to the lesion location and individual principle,we determined CTV to PTV margins in the Treatment Planning System.2 This study we selected formula 2.5 ? + 0.7 ? to calculate the setup margins of different groups of lung cancer patients.We placed system error(?)and random error(?)into formula,to observe the differences of margins between different groups of lung cancer patients.Results:1 In automatic method,the setup margins of central type in X?Y?Z axis are 3.21?7.04?4.62 mm.The setup margins of peripheral type in X?Y?Z axis are 3.54?9.98?6.26 mm.The setup margins of left lung type in X?Y?Z axis are 2.79?9.53?5.31 mm.The setup margins of right lung type in X?Y?Z axis are 3.05?7.04?4.62 mm.The setup margins of upper lobe type in X?Y?Z axis are 2.36?6.15?4.89 mm.The setup margins of lower lobe type in X?Y?Z axis are 4.02?10.46?7.82mm;2 In manual method,the setup margins of central type in X?Y?Z axis are2.69?9.58?6.31 mm.The setup margins of peripheral type in X?Y?Z axis are3.96?9.21?8.15 mm.The setup margins of left lung type in X?Y?Z axis are3.30?8.32?7.41 mm.The setup margins of right lung type in X?Y?Z axis are3.75?8.44?6.81 mm.The setup margins of upper lobe type in X?Y?Z axis are2.68?7.23?6.74 mm.The setup margins of lower lobe type in X?Y?Z axis are4.68?10.74?9.75mm;3 The setup margins of peripheral type is larger than central type,the setup margins of lower lobe type is upper lobe type;4 There were no differences between the margins of automatic method and manual method.Conclusions: Through the observation of 46 cases of lung cancer patients with CBCT scanning to obtain the system error and random error,and then substituted into the formula 2.5 ? + 0.7 ? we found,no matter the automatic registration method or the manual registration method,the difference of setup margins is not big.It is suggested that the automatic registration method is used for the registration of CBCT and CT in the patients with lung cancer. |
PDF Full Text Request