Objective:To investigate the role of three-dimensional dose distribution-based deep learning model in predicting distant metastasis of head and neck cancer.Methods:Radiotherapy and clinical follow-up data of 237 patients with head and neck cancer undergoing intensity-modulated radiotherapy (IMRT) from 4 different institutions were collected. Among them, 131 patients from HGJ and CHUS institutions were used as the training set, 65 patients from CHUM institution as the validation set, and 41 patients from HMR institution as the test set. Three-dimensional dose distribution and GTV contours of 131 patients in the training set were input into the DM-DOSE model for training and then validated with validation set data. Finally, the independent test set data were used for evaluation. The evaluation content included the area under receiver operating characteristic curve (AUC), balanced accuracy, sensitivity, specificity, concordance index and Kaplan-Meier survival curve analysis.Results:In terms of prognostic prediction of distant metastasis of head and neck cancer, the DM-DOSE model based on three-dimensional dose distribution and GTV contours achieved the optimal prognostic prediction performance, with an AUC of 0.924, and could significantly distinguish patients with high and low risk of distant metastasis (log-rank test, P<0.001). Conclusion:Three-dimensional dose distribution has good predictive value for distant metastasis in head and neck cancer patients treated with IMRT, and the constructed prediction model can effectively predict distant metastasis.

Objective:To establish an automatic segmentation network based on different receptive fields for gross target volume (GTV) and organs at risk in patients with nasopharyngeal carcinoma.Methods:Radiotherapy data of 100 cases of nasopharyngeal carcinoma including CT images and GTV and organs at risk delineated by the physicians were collected. Ninety plans were randomly selected as the training dataset, and the other 10 plans as the validation dataset. Firstly, the images were subject to three data augmentation methods including center cropping, vertical flipping and rotation (-30°to 30°), and then input into MA_net networks proposed in this study for training. The model performance of networks was assessed by the number of network parameters (NP), floating-point number (FPN), the running memory (RM) and Dice index (DI), and eventually compared with DeeplabV3+ , PSP_net, UNet+ + and U_Net networks.Results:When the input image was in the size of 240×240, MA_net had a NP of 23.20%, 20.10%, 25.55% and 27.11% of these 4 networks, 50.02%, 19.86%, 6.37% and 13.44% for the FPN, 40.63%, 23.60%, 11.58% and 14.99% for the RM, respectively. For the DI of GTV, MA_net was 1.16%, 2.28%, 1.27% and 3.59% higher than these 4 networks. For the average DI of GTV and OAR, MA_net was 0.16%, 1.37%, 0.30% and 0.97% higher than these 4 networks.Conclusion:Compared with those four networks, the proposed MA_net network has slightly higher Dice index with fewer parameters, lower FPN and smaller RM.

In order to suppress the geometrical artifacts caused by random jitter in ray source scanning, and to achieve flexible ray source scanning trajectory and meet the requirements of task-driven scanning imaging, a method of free trajectory cone-beam computed tomography (CBCT) reconstruction is proposed in this paper. This method proposed a geometric calibration method of two-dimensional plane. Based on this method, the geometric calibration phantom and the imaging object could be simultaneously imaged. Then, the geometric parameters could be obtained by online calibration method, and then combined with the geometric parameters, the alternating direction multiplier method (ADMM) was used for image iterative reconstruction. Experimental results showed that this method obtained high quality reconstruction image with high contrast and clear feature edge. The root mean square errors (RMSE) of the simulation results were rather small, and the structural similarity (SSIM) values were all above 0.99. The experimental results showed that it had lower image information entropy (IE) and higher contrast noise ratio (CNR). This method provides some practical value for CBCT to realize trajectory freedom and obtain high quality reconstructed image.
Algorithms ; Calibration ; Cone-Beam Computed Tomography ; Image Processing, Computer-Assisted ; Phantoms, Imaging

Objective:To establish a correlation model between MRI and CT images to generate synthetic-CT (sCT) of head and neck cancer during MRI-guided radiotherapy by using generative adversarial networks (GAN).Methods:Images and IMRT plans of 45 patients with nasopharyngeal carcinoma were collected before treatment. Firstly, the MRI (T1) and CT images were preprocessed, including rigid registration, clipping, background removal and data enhancement, etc. Secondly, the cases were trained by GAN, of which 30 cases were randomly selected and put into the network as training set images for modeling and learning, and the other 15 cases were used for testing. The image quality of predicted sCT and real CT were statistically compared, and the dose distribution recalculated upon predicted sCT was statistically compared with that of real planned dose distribution.Results:The mean absolute error of the predicted sCT of the testing set was (79.15±11.37) HU, and the SSIM value was 0.83±0.03. The MAE values of dose distribution difference at different regional levels were less than 1% compared to the prescription dose. The gamma passing rate of the sCT dose distribution was higher than 92% and 98% under the 2mm/2% and 3mm/3% criteria.Conclusions:We have successfully proposed and realized the generation of sCT for head and neck cancer using GAN, which lays a foundation for the implementation of MRI-guided radiotherapy. The comparison of image quality and dosimetry shows the feasibility and accuracy of this method.

Objective:To compare the accuracy and generalized robustness of three predictive models of knowledge-based treatment strategies for radiotherapy for optimized model selection.Methods:The clinical radiotherapy plans of 45 prostate cancer (PC) cases and 25 nasopharyngeal cancer (NPC) cases were collected, and analyzed using three models (Z, L and S model), proposed by Zhu et al, Appenzoller et al and Shiraishi et al, respectively, to predict the dose-volume histogram (DVH) of bladder and rectum on PC cases and that of left and right parotid on NPC cases. The prediction error was measured by the difference of area under the predicted DVH and the clinical DVH curves (|V (pre_DVH)-V (clin_DVH)|), where a smaller prediction error implies a greater prediction accuracy. The accuracies of these three models were compared on the single organ at risk (OAR), and the generalized robustness of models was evaluated and compared by calculating the standard deviation of the prediction accuracy on different OAR. Results:For bladder and rectum, the prediction error of L model (0.114 and 0.163, respectively) was significantly higher than those values of Z and S models (≤0.071, P<0.05); for left parotid gland, the predicted error of S model (0.033) did not present significant difference from those values of Z and L models (≤0.025, P>0.05); for right parotid gland, S model (0.033) demonstrated significantly higher prediction error than those of Z and L models (≤0.028, P<0.05). Regarding different OAR, S model showed a lower standard deviation of prediction accuracy when comparing to Z and L models (0.016, 0.018 and 0.060, respectively). Conclusions:In the prediction of DVH in bladder and rectum of PC, Z and S models were more accurate than L model. In contrast, Z and L models demonstrated higher accuracy than S model in the prediction of left and right parotid glands of NPC. In respect to different OAR, the generalized robustness of S model was superior than the other two models.

Objective:To explore the value of prenatal MRI in the diagnosis of isolated mild and moderate bilateral ventriculomegaly and neural development of the fetuses after birth.Methods:This is a retrospective study involving 244 singleton fetuses with isolated mild or moderate lateral ventriculomegaly diagnosed by both prenatal ultrasound and MRI in Huzhou Maternity & Child Health Care from May 2013 to June 2017, consisting of 82 cases with bilateral ventriculomegaly (BVM) and 162 with unilateral ventriculomegaly (UVM). The two groups were further divided into two subgroups: mild (lateral ventricle width: 10.0-12.0 mm, bilateral 56 cases, unilateral 120 cases) and moderate group (lateral ventricle width: >12.0-<15.0 mm, bilateral 26 cases, unilateral 42 cases). In addition, 50 singleton fetuses without any abnormality in the nervous system in prenatal check were included in the control group during the same period. All neonates were reexamined by ultrasound within one week after birth, and followed up regularly at the age of 3, 6, 12 and 18 months. Gesell Development Schedules (GDS) were used to evaluate the central nervous system's function, and postnatal changes in lateral ventriculomegaly were observed. Statistical analysis was performed by t, F, Chi-square tests (or Fisher's exact test). Results:(1) There was no difference among intervals between MRI scan and delivery in the BVM, UVM, and the control groups. The disappearance rate of lateral ventriculomegaly after birth was 80.4% (45/56) in the mild BVM group, 42.3% (11/26) in the moderate BVM group, 88.3% (106/120) in the mild UVM group, and 57.1% (24/42) in the moderate UVM group ( χ2=35.183, P<0.001). (2) The GDS evaluation results in the BVM group at 6, 12, and 18 months after birth were worse than those in the UVM group (all P<0.0167). The GDS evaluation results in the BVM group were worse than those in the control group at 3 and 6 months after birth [3 months: normal: 58.5% (48/82) vs 86.0% (43/50), borderline: 22.0% (18/82) vs 10.0% (5/50), delay: 19.5% (16/82) vs 4.0% (2/50), χ2=11.425; 6 months: normal: 63.4% (52/82) vs 88.0% (44/50), borderline: 19.5% (16/82) vs 8.0% (4/50), delay: 17.1% (14/82) vs 4.0% (2/50), χ2=9.678; all P<0.0167]. (3) The GDS evaluation results in the moderate BVM group at 6, 12, and 18 months after birth were worse than those in the moderate UVM group [6 months: normal: 30.8% (8/26) vs 69.0% (29/42), borderline: 30.8% (8/26) vs 21.4% (9/42), delay: 38.5% (10/26) vs 9.5% (4/42), χ2=11.417; 12 months: normal: 53.8% (14/26) vs 88.1% (37/42), borderline: 23.1% (6/26) vs 9.5% (4/42), delay: 23.1% (6/26) vs 2.4% (1/42), χ2=11.199; 18 months: normal: 65.4% (17/26) vs 95.2% (40/42), borderline: 15.4% (4/26) vs 2.4% (1/42), delay: 19.2% (5/26) vs 2.4% (1/42), χ2=10.568; all P<0.0167]. The GDS evaluation results of the moderate BVM group at 3, 6, 12, and 18 months after birth were worse than the control group. (4) In the BVM group, the GDS scores at 18 months of age were better than those at three months of age ( χ2=8.224, P=0.016). Conclusions:(1) Most mild BVM would disappear spontaneously after birth, while more in mild UVM cases. (2) The postnatal GDS evaluation results of the BVM group is significantly worse than that of the UBM group at months of age; (3) Fetuses with less severe isolated BVM are more likely to have improved GDS score after birth.

Objective To propose a treatment planning optimization algorithm which can make full use of OAR dose distribution prediction meanwhile improving the output planning quality as much as possible.Methods We had reformulated an FMO function under the guidance of dose distribution prediction and also integrated equivalent uniform dose (gEUD) based on the consideration of prediction uncertainty,for providing optimal solution.Performance of the method was evaluated by comparing the optimized IMRT plan quality of 8 cervical cancers in the term of DVH curves,dose distribution and dosimetric endpoints with the original ones.Results The proposed method had a feasible,fast solution.Compared with original plan,its output plan had better plan quality in better dose homogeneity,less hot spot and further dose sparing for OARs.V30,V45 of rectum was decreased by (6.60±3.53)％ and (17.03±7.44)％,respectively,with the statistically significant difference (t=-4.954,-6.055,P＜0.05).V30,V45 of bladder was decreased by (14.74 ± 5.61) ％ and (14.99 ± 4.53) ％,respectively,with the statistically significant difference (t=-6.945,-8.759,P＜0.05).Conclusions We have successfully developed a predicted dose distribution and equivalent uniform dose-based planning optimization method,which is able to make good use of 3D dose prediction and ensure the output plan quality for intensity modulated radiation therapy.

Objective To establish a three-dimensional (3D) dose prediction model,which can predict multiple organs simultaneously in a single model and automatically learn the effect of the geometric anatomical structure on dose distribution.Methods Clinical radiotherapy plans of patients diagnosed with the same type of tumors were collected and retrospectively analyzed.For every plan,each organs at risk (OAR) voxel was regarded as the study sample and its deposited dose was considered as the dosimetric feature.A regularized multi-task learning method than could learn the relationship among different tasks was employed to establish the relationship matrix among tasks and the correlation between geometric structure and dose distribution among organs.In this experiment,the spinal cord,brainstem and bilateral parotids involved in the intensity-modulated radiotherapy (IMRT) plan of 15 nasopharyngeal cancer patients were utilized to establish the multi-organ prediction model.The relative percentage error between the predicted dose of voxel and the clinical planning dose was calculated to assess the feasibility of the model.Results Ten cases receiving IMRT plans were utilized as the training data,and the remaining five cases were used as the test data.The test results demonstrated a higher prediction accuracy and less data demand.And the average voxel dose errors among the spinal cord,brainstem and the left and right parotids were (2.01±0.02)％,(2.65± 0.02) ％,(2.45± 0.02) ％ and (2.55± 0.02) ％,respectively.Conclusion The proposed model can accurately predict the dose of multiple organs in a single model and avoid the establishment of multiple single-organ prediction models,laying a solid foundation for patient-specific plan quality control and knowledge-based treatment planning.

Objective To explore the value of prenatal MRI in the diagnosis of fetal simple expansion of lateral ventricle(ventriculomegaly), and follow up the nervous system development status after birth. Methods Simple expansion of the lateral ventricle fetus by prenatal MRI examination were collected in Huzhou Maternal and Child Care Hospital from May 2013 to June 2015, 126 cases of live births in expansion group, 50 normal cases were recruited in the same period as the control group. In expansion group, fetal subgroup analysis was done:(1) unilateral or bilateral lateral ventricle expasion:one group was 98 cases was lateral ventricle expansion (77.8%, 98/126), expansion of bilateral ventricle group was 28 cases (22.2%, 28/126). (2) Prenatal MRI in the diagnosis of the lateral ventricle of expansion: expansion of the lateral ventricle width was greater than 10.0 mm, if both sides were expanding, the expand width was the heavier one side, divided into 3 subgroups: ①Expansion in group A (lateral ventricle width 10.0-12.0 mm) were 88 cases (69.8%, 88/126).②Expansion in group B (lateral ventricle width 12.1-15.0 mm) were 29 cases (23.0%, 29/126). ③Expansion of group C (lateral ventricle width> 15.0 mm) were 9 cases (7.12%, 9/126). All 176 cases were followed up after birth at the 3rd, 6th, 12th, 18th month (corrected age was used for premature babies), and Gesell developmental schedules (GDS) were used to evaluate the neurobehavioral development. Results (1) The MRI results after birth:21 cases were followed up by MRI after birth. In group A, 11 cases had MRI and 9 were normal (the ventricular width<10.0 mm after birth) , the other 2 cases were stable (the ventricular width measured first time after birth was ≥10.0 mm, but the difference was within 2.0 mm from the MRI before birth). In group B, 4 cases had MRI, 1 was normal, 1 was stable, and 2 cases were getting better (the ventricular width measured first time after birth was ≥10.0 mm, but the width decreased more than 2.0 mm from the MRI before birth). In group C, 6 cases had MRI. 3 cases were getting better and 3 cases were stable. (2) Overall GDS results:expansion group after the birth of the 3rd, 6th, 12th, 18th month GDS evaluation results compared with control group, respectively, the differences were not statistically significant (all P>0.05). (3) The GDS results among the subgroups:in each evaluation after birth, there were no statistically significant differences between group A and the control group (all P>0.05). The GDS results of group B at the 3rd and 6th month were lower than those of the control group (P<0.05); while there were no statistically significant differences between the 2 goups at the 12th and 18th month (P>0.05). And for group C, statistically significant differences were found compared to the control group at each follow-up time (all P<0.05). (4) GDS results at different times after birth in the expansion group:there was no statistically significant difference between the results at the 3rd and 6th month (P>0.05). But when the result at the 3rd month was compared to the results of the 12th or 18th month, the differences were statistically significant (P<0.05). GDS result of 6th months after birth compared with 12th and 18th months, respectively, there were no statistically significant differences (P>0.05). There was no statistically significant difference between the results at the 12th and 18th month (P>0.05). (5) The GDS results in unilateral and bilateral ventricle expansion:at the 18th month, among the 98 unilateral cases, 86 (87.8%, 86/98) had normal GDS results(>85 scores);8 (8.2%, 8/98) had borderline results (75-85 scores);4 (4.1%, 4/98) had delayed results (<75 scores). Among the 28 bilateral cases, 23 (82.1%, 23/28) had normal GDS results;3 (10.7%, 3/28) had borderline results; 2 (7.1%, 2/28) had delayed results. There was no statistically significant difference (P>0.05). Conclusions Among the simple expansion of lateral ventricle, those whose ventricular width are≤12.0 mm may not need clinical treatment. If the width is between 12.1 to 15.0 mm, closely follow-up and targeted rehabilitation training after birth are recommended. When the width is more than 15.0 mm, the risk of the central nervous system function delay is significantly increased, and early intervention might improve the prognosis.

Objective To investigate the clinical value of prenatal MRI in the diagnosis and differential diagnosis of congenital bronchopulmonary sequestration (BPS). Methods From January 2009 to December 2014, 16 fetuses with BPS were diagnosed by fetal MRI in Huzhou Maternity and Child Care Hospital and Shanghai Children′s Medical Center Affiliated to Shanghai Jiaotong University School of Medicine. The clinical data of these cases were analyzed retrospectively. All were singleton pregnancy, and MRI was carried out within 24-48 hours after routine prenatal ultrasound. All the neonates underwent postnatal enhanced CT scan or surgical biopsy after birth, and the results were compared to prenatal MRI diagnosis. Results (1)With prenatal MRI, 16 cases were diagnosed BPS. The lesions located in left lung in 10 cases, and right lung in 6 cases. As the scope of the lesion, 3 cases located in the whole left lung, 6 cases limited to the left lower lobe, and 1 case was subdiaphragmatic on the left side. 2 cases located in the whole right lung and 4 cases limited to the right lower lobe. One case complicated oligoamnios, and one had pleural effusion. Supplying vessels could be found in 14 cases.(2)When the postnatal results were compared with prenatal MRI, 15 cases were comfirmed as BPS (15/16), including 10 intralobar cases 5 extralobar cases. One that was diagnosed as BPS by prenatal MRI was confirmed to be congenital cystic adenomatoid malformation (CCAM) by pathology. The accuracy of prenatal MRI diagnosis of BPS was 15/16. Prenatal ultrasound missed one case and misdiagnosed two cases, as one was mistakened as CCAM and the other as cystic teratoma. Conclusion Prenatal MRI has good clinical value in the diagnosis and differential diagnosis of fetal BPS.