Objective To test intraoperative radiotherapy with mobile photon beam using the INTRABEAM system ( Germany) , and to analyze the dosimetric characteristics of low?energy photon beam using X?ray source and spherical applicators and explore its potential limitations in clinical application. Methods A special water phantom, a parallel?plate ionization chamber, and an electrometer were used to measure the depth dose rates and isotropy of dose distribution in x/y plane of X?ray source and different spherical applicators in the INTRABEAM system. Those data were then compared with the system data. Results For the X?ray source, the deviation of observed depth dose rate and isotropy in the x/y plane from the system data were-2.16%± 1. 36% and-1.9%~ 2. 1%, respectively. For applicators with different diameters, the deviation of observed depth dose rate, transfer coefficient, and isotropy in x/y plane from the system data were-10.0%~2. 3%,-8.9%~4. 2%, and-1.6%~2. 6%, respectively. Surface dose rate and dose gradient became larger with the decrease in the diameter of the spherical applicator. The measurement of depth dose rate and isotropy of X?ray source and spherical applicators showed good repeatability. The influencing factors for measurement accuracy included the positioning error of ionization chamber, energy response, noise current, and correction factor f ’ ( R ) . Conclusions This study reveals the dosimetric characteristics of the INTRABEAM system, verifies the accuracy of the system data, and obtains the data for clinical application and routine quality assurance. However, large dose gradient and small therapeutic range may limit its wide clinical application.

[Objective]To compare and contrast the dosimetry between Tomo planning and BrainLab planning for lung metasta-ses in stereotactic body radiation therapy(SBRT).[Methods]Four Patients with one,two,three and four metastases were selected. The PTV is 2.89 ± 1.15 cm3. Two plannings with total dose of 50 Gy to cover 95% of PTV ,5 Gy/Fraction and 10 fractions were designed using Tomo planning system and BrainLab planning system respectively. The DVH curves of spinal cord ,both lungs and normal tissue were compared. The conformity index andhomogeneityindex were analyzed as well.[Results]The homogeneity index (HI)and conformity index(CI)of the targets in Tomo planning system were 1.0314 ± 0.0700 and 0.687 ± 0.075,respectively. In BrainLab planning system the HI and CI of the targets were 1.0764 ± 0.1241 and 0.571 ± 0.042,respectively. To HI the P value in T test was less than 0.01 and the HI was better in Tomo than BrainLab and so was CI. The dose to spinal cord was higher in BrainLab planning system than that in Tomo. The dose to nomal tissue and both lungs were not different in the two planning systems and V20 of lung is as small as 10%.[Conclusions]For small volume lung metastases which longest diameter were less than 4 cm,the tomotherapy should be better choice.

Objective To study the mathematical predicting model of parotid DVH for the NPC IMRT planning, and its accuracy with the analysis of medical data. Methods 50 NPC radiotherapy treatment plans with same beam setup were chosen as sample data set, then their parotid DVHs and distance of voxels in the parotid to the target volumes were calculated with self-developed program to form the distance to target histogram ( DTHs);principal component analysis was applied to DVHs and DTHs to acquire their principal components ( PCs) ,and then nonlinear multiple variable regression was used to model correlation between the DTHs' PCs, parotids volume, PTVs and the DVHs. Another 10 plans were chosen as test data set to evaluate the efficacy and accuracy of the final model by comparing the DVHs calculated from our model with those calculated from the TPS. Results Up to 97% information of DTHs and DVHs can be represented with 2 to 3 components, the average fitting error of sample data set was (0±3. 5)%;in the 10 test cases, the shapes of DVH curves calculated from predicting model was highly the same with those from the TPS, the average modeling error was (-0.7± 4. 4)%,the accuracy of prediction was up 95%. Conclusions Our developed model can be used as a quality evaluating tool to predict and assure the dose distribution in parotid of NPC radiotherapy treatment planning effectively and accurately.

Objective To evaluate the impacts of slice thickness and registration frame range on the accuracy of cone-beam computed tomography ( CBCT) image-guided head and neck ( HN) radiotherapy, and to provide a basis for positioning correction in image-guided radiotherapy.Methods A planned CT scan was performed for an anthropomorphic HN phantom with slice thickness of 1 mm and 3 mm and simulated positioning errors in x, y, and z directions on the accelerator.CBCT scan and reconstruction were performed with slice thickness of 1 mm and 3 mm.Two different registration frame ranges were used ( range 1:from C7 to superior orbit;range 2:from C7 to calvaria ) .Automatic bony registration was performed for CBCT and planned CT images with slice thickness of 1 mm and 3 mm.The registration accuracy was evaluated.Results For range 1, the registration errors in x, y, and z directions with a slice thickness of 1 mm were significantly lower than those with a slice thickness of 3 mm (0.5±0.2 mm vs.-0.7±0.2 mm, P=0.00;0.5±0.3 mm vs. 1.0±0.3 mm, P=0.00;-0.1±0.5 mm vs.1.5±0.5 mm, P=0.00).For range 2, the registration errors in x, y, and z directions with a slice thickness of 3 mm were-0.4±0.2 mm, 0.5±0.2 mm, and 0.7±0.4 mm, respectively.Conclusions Engagement of calvaria in registration range can substantially enhance the registration accuracy in CBCT or CT images for HN.The registration error with slice thickness of 1 mm can be controlled within 1 mm.

Objective To evaluate the accuracy of image registration based on bony structure (RBS) and grey-scale (RGS) in positioning correction of radiation treatment,and their reliability in clinical application.Methods Setup errors of anthropomorphic phantom (chest& abdomen,head& neck) were simulated with x-,y-,z-directions.CBCT images were acquired for each simulation and registered with planning CT.using bony structure and grey-scale registration separately.Geometry accuracy of all registration were then obtained and analyzed.Results The errors of RBS and RGS in x-,y-,z-directions were (-0.65 ±0.22) mm and (-0.70±0.17) mm (P=0.00),(1.02 ±0.27) mm and (0.90±0.20) mm (P =0.00),(1.46 ± 0.53) mm and (1.47 ± 0.47) mm (P =0.54) for head& neck positioning; with (0.82±0.33) mm and (0.79±0.18) mm (P=0.03),(2.45±1.17) mm and (1.61 ±0.84) mm (P =0.00),(1.44 ± 3.25) mm and (0.19 ± 1.11) mm (P =0.00) for chest& abdomen positioning.Conclusions RGS is more accurate and stable than RBS.The accuracy of image registration is a little better for head& neck than that for chest& abdomen.The algorithms of image registration used in clinical application needs to be tested independently and the systematic error needs to be corrected before applying in different treatment techniques according to their accuracy requirement.

BACKGROUND: Ultrasonic coupling agent is indispensable in ultrasonic imaging. The service life and change frequency of the ultrasonic probe is determined by the internal coupling agent of the chosen hydrogel material. Due to defects in the existing gel-formula, after a period of time in use it can cause the surface depression of probes and thus influence the ultrasonic imaging quality. OBJECTIVE: To optimize the gel formula of the coupling agent. METHODS: The amplified synthesis process of polyacrylamide hydrogel was optimized by changing the ratio of monomer concentration to N, N-methylenebisacrylamide crosslinker in the synthesis conditions, and the factors influencing its acoustic properties were explored to meet the technical requirements in the clinical use process. RESULTS AND CONCLUSION: When the monomer concentration was set to 70 g/L and the mass ratio of monomer to crosslinker was 7:1, the polyacrylamide hydrogel was in the best state. The obtained polyacrylamide hydrogel had a stable swelling rate, had a strong ability to keep water and make the volume reduce under the influence of the operating temperature of ultrasound, and its relative amplitude changed less with the change of ultrasonic frequency. The sound attenuation coefficient of polyacrylamide hydrogel would be higher when the frequency of ultrasonic sound source was higher. The changes of monomer concentration would have significant impacts on the acoustic properties of polyacrylamide hydrogel. Therefore, by controlling the monomer concentration ratio, we can prepare the coupling agent that exhibits relatively stable acoustic properties at an optimum concentration and achieves an optimal imaging effect.

Objective To construct and investigate the multi-leaf collimator (MLC) fault prediction model of Varian NovalisTx medical linear accelerator based on BP neural network.Methods The MLC fault data applied in clinical trial for 18 months were collected and analyzed.The total use time of accelerator,the quantity of patients per month,average daily working hours of accelerator,volume of RapidArc plans and time interval between accelerator maintenance were used as the input factors and the prediction of MLC fault frequency was considered as the output result.The BP neural network model of MLC fault prediction was realized by AMORE package of R language and the simulation results were validated.Results The model contained 3 layers of network to realize the input-output switch.There were 5 nodes in the input layer,13 nodes in the hide layer and 1 node in the output layer,respectively.The transfer function from the input layer to the hide layer selected the tansig function and purelin function was used from the hide layer to the output layer.The maximum time of training was pre-set as 150 in the designed model.Actually,111 times of training were performed.The pre-set error was 3％ and the actual error was 2.7％,which indicated good convergence.The simulation results of MLC fault applied in clinical trial for 18 months were similar to the actual data.Conclusions The BP neural network model realized by R language of MLC fault prediction can describe the mapping relationship between fault factors and fault frequency,which provides references for the understanding of accelerator fault and management of spare parts inventory.