Objective:To investigate the current application of blood purification in the treatment of acute poisoning within Jiangsu Province and to evaluate the impact of extracorporeal blood purification on the clinical outcomes of critically poisoned patients.Methods:This multicenter, cross-sectional real-world observational study followed patients presenting with poisoning to the emergency departments of nine hospitals in Jiangsu Province between June 2015 and May 2019. Data were collected on demographic characteristics, vital signs within the first hour of emergency presentation, treatment modalities, length of hospital stay, and survival outcomes. Clinical data from patients who underwent extracorporeal blood purification were compared with those who did not, using the Wilcoxon rank-sum test and Chi-square test.Results:A total of 4 178 poisoning cases were included between June 2015 and May 2019. Among them, 21.7% (908/4 178) received blood purification therapy, while 78.3% (3 270/4 178) did not. Hemoperfusion (90.4%) was the most frequently employed method, followed by continuous renal replacement therapy (CRRT) (4.4%). In combined blood purification modalities, 4.8% underwent hemoperfusion combined with CRRT, 0.1% received hemoperfusion with plasma exchange, and another 0.1% underwent hemoperfusion combined with both CRRT and plasma exchange. Among patients who underwent blood purification, pesticide poisoning was the most prevalent (76.3%), with the most common toxic agents being paraquat (23.7%), dichlorvos (8.7%), methamidophos (5.2%), omethoate (4.0%), and glyphosate (3.7%). Compared to the non-blood purification group, patients in the blood purification group were more likely to present within the first hour with a low Glasgow Coma Scale (GCS) score (3-8) (22.6% vs. 9.7%, P <0.05), low mean arterial pressure (8.0% vs. 3.2%, P <0.05), longer hospital stays [5(3,9) days vs. 2(1,4) days, P <0.05] and a higher in-hospital mortality rate (21.1% vs. 5.3%, P <0.05). Follow-up via telephone 28 days after discharge revealed a survival rate of 78.9%, with a mortality rate of 21.1% in the blood purification group. Conclusions:Hemoperfusion is the most commonly utilized blood purification technique for treating poisoning in Jiangsu Province, with pesticides being the primary toxic agents treated. Although the mortality rate is higher in the blood purification group, the intervention may still contribute to improved patient outcomes.

Artificial intelligence (AI) technologies are being widely applied in radiotherapy. However, the integration of AI into clinical workflows of radiotherapy faces a series of challenges, such as poor model interpretability, domain shifts between clinical application and training data, and the inherent model uncertainties. Therefore, case-specific quality assurance (QA) is essential before deploying AI models in clinical practice. This paper reviews and summarizes QA methodologies for the application of AI models in radiotherapy across four key areas: image registration, image generation, region of interest segmentation, and treatment planning.

Objective:To evaluate the feasibility of using the domestic ArcherQA system for fast and simplified independent verification of CyberKnife (CK) stereotactic radiotherapy (SRT) plans.Methods:SRT plans of 57 patients treated with CK at Tianjin Medical University Cancer Institute and Hospital from August 2021 to August 2022 were retrospectively analyzed, including 15 intracranial, 30 pulmonary, and 12 abdominal tumors cases. Point-dose and planar-dose verifications were performed using an ionization chamber and radiochromic films embedded in a homogeneous phantom, and the results were compared with those calculated by the treatment planning system (TPS). The localization CT images and corresponding SRT plans were imported into the ArcherQA system for independent dose verification and analysis. The correlation between ArcherQA results and phantom measurements was analyzed, with comparisons of target mean dose differences and γ pass rates.Results:Phantom measurement results showed, the measured point-dose differences for intracranial, lung, and abdominal plans were -0.94% ± 3.22%, 1.92% ± 2.05%, and 2.12% ± 0.77%, respectively. The mean dose differences in target dose calculation between ArcherQA and TPS: intracranial in the gross tumor volume (GTV) regions were 0.34% ± 2.21%, lung tumor GTV were -2.47% ± 2.46%, and abdominal tumor GTV were 0.80% ± 2.61%, respectively. Among them, the abdominal GTV region showed the highest correlation between ArcherQA and measured results ( r=0.78). The average two-dimensional γ pass rates (2 mm/2%, threshold=10%) measured using phantom films were 95.92% ± 2.35% for intracranial, 95.70% ± 2.74% for lung, and 96.74% ± 3.41% for abdominal tumors plans, respectively. The three-dimensional ArcherQA results showed comparable γ pass rates (1 mm/2%, threshold=10%) for lung and abdominal GTV and PTV regions, with similar medians and data dispersion to film measurements. Conclusions:The ArcherQA system enables rapid and efficient independent dose verification of CK SRT plans without the need for additional hardware. The verification results show good correlation with phantom measurements, supporting its potential as an auxiliary quality assurance tool in clinical CK SRT implementation.

Artificial intelligence (AI) technologies are being widely applied in radiotherapy. However, the integration of AI into clinical workflows of radiotherapy faces a series of challenges, such as poor model interpretability, domain shifts between clinical application and training data, and the inherent model uncertainties. Therefore, case-specific quality assurance (QA) is essential before deploying AI models in clinical practice. This paper reviews and summarizes QA methodologies for the application of AI models in radiotherapy across four key areas: image registration, image generation, region of interest segmentation, and treatment planning.

Objective:To evaluate the feasibility of using the domestic ArcherQA system for fast and simplified independent verification of CyberKnife (CK) stereotactic radiotherapy (SRT) plans.Methods:SRT plans of 57 patients treated with CK at Tianjin Medical University Cancer Institute and Hospital from August 2021 to August 2022 were retrospectively analyzed, including 15 intracranial, 30 pulmonary, and 12 abdominal tumors cases. Point-dose and planar-dose verifications were performed using an ionization chamber and radiochromic films embedded in a homogeneous phantom, and the results were compared with those calculated by the treatment planning system (TPS). The localization CT images and corresponding SRT plans were imported into the ArcherQA system for independent dose verification and analysis. The correlation between ArcherQA results and phantom measurements was analyzed, with comparisons of target mean dose differences and γ pass rates.Results:Phantom measurement results showed, the measured point-dose differences for intracranial, lung, and abdominal plans were -0.94% ± 3.22%, 1.92% ± 2.05%, and 2.12% ± 0.77%, respectively. The mean dose differences in target dose calculation between ArcherQA and TPS: intracranial in the gross tumor volume (GTV) regions were 0.34% ± 2.21%, lung tumor GTV were -2.47% ± 2.46%, and abdominal tumor GTV were 0.80% ± 2.61%, respectively. Among them, the abdominal GTV region showed the highest correlation between ArcherQA and measured results ( r=0.78). The average two-dimensional γ pass rates (2 mm/2%, threshold=10%) measured using phantom films were 95.92% ± 2.35% for intracranial, 95.70% ± 2.74% for lung, and 96.74% ± 3.41% for abdominal tumors plans, respectively. The three-dimensional ArcherQA results showed comparable γ pass rates (1 mm/2%, threshold=10%) for lung and abdominal GTV and PTV regions, with similar medians and data dispersion to film measurements. Conclusions:The ArcherQA system enables rapid and efficient independent dose verification of CK SRT plans without the need for additional hardware. The verification results show good correlation with phantom measurements, supporting its potential as an auxiliary quality assurance tool in clinical CK SRT implementation.

Objective:To investigate the safety and therapeutic effect of split liver transplantation (SLT) in clinical application.Methods:This is a retrospective case-series study. The clinical data of 203 consecutive SLT, 79 living donor liver transplantation (LDLT) and 1 298 whole liver transplantation (WLT) performed at the Third Affiliated Hospital of Sun Yat-sen University from July 2014 to July 2023 were retrospectively analyzed. Two hundred and three SLT liver grafts were obtained from 109 donors. One hundred and twenty-seven grafts were generated by in vitro splitting and 76 grafts were generated by in vivo splitting. There were 90 adult recipients and 113 pediatric recipients. According to time, SLT patients were divided into two groups: the early SLT group (40 cases, from July 2014 to December 2017) and the mature SLT technology group (163 cases, from January 2018 to July 2023). The survival of each group was analyzed and the main factors affecting the survival rate of SLT were analyzed. The Kaplan-Meier method and Log-rank test were used for survival analysis.Results:The cumulative survival rates at 1-, 3-, and 5-year were 74.58%, 71.47%, and 71.47% in the early SLT group, and 88.03%, 87.23%, and 87.23% in the mature SLT group, respectively. Survival rates in the mature SLT group were significantly higher than those in the early SLT group ( χ2=5.560, P=0.018). The cumulative survival rates at 1-, 3- and 5-year were 93.41%, 93.41%, 89.95% in the LDLT group and 87.38%, 81.98%, 77.04% in the WLT group, respectively. There was no significant difference among the mature SLT group, the LDLT group and the WLT group ( χ2=4.016, P=0.134). Abdominal hemorrhage, infection, primary liver graft nonfunction,and portal vein thrombosis were the main causes of early postoperative death. Conclusion:SLT can achieve results comparable to those of WLT and LDLT in mature technology liver transplant centers, but it needs to go through a certain time learning curve.

Objective To improve the performance of auto-segmentation of prostate target area and organs-at-risk in online magnetic resonance image and enhance the efficiency of magnetic resonance imaging-guided adaptive radiotherapy(MRIgART)for prostate cancer.Methods A retrospective study was conducted on 40 patients who underwent MRIgART for prostate cancer,including 25 in the training set,5 in the validation set,and 10 in the test set.The planning CT images and corresponding contours,along with online MR images,were registered and input into a deep learning network for online MR image auto-segmentation.The proposed method was compared with deformable image registration(DIR)method and single-MR-input deep learning(SIDL)method.Results The overall accuracy of the proposed method for auto-segmentation was superior to those of DIR and SIDL methods,with average Dice similarity coefficients of 0.896 for clinical target volume,0.941 for bladder,0.840 for rectum,0.943 for left femoral head and 0.940 for right femoral head,respectively.Conclusion The proposed method can effectively improve the accuracy and efficiency of auto-segmentation in MRIgART for prostate cancer.

Objective To evaluate the effects of the multiple factors especially image geometric accuracy of the imaging system on the segmentations of target areas and organs-at-risk.Methods The study used phantoms to test the imaging performance of the 1.5T magnetic resonance(MR)linear accelerator system,including the assessments of MR image geometric distortion and the segmentation errors caused by factors such as image geometric distortion.Model 604-GS large field MR image distortion phantom was used to explore the geometric distortion of the MR images for MR-guided radiotherapy;and CIRS Model 008z upper abdominal phantom was used to analyze the segmentation errors of target areas and organs-at-risk.Results The average geometric distortion and maximum distortion of 3D T1WI-FFE images vs 3D T2WI-TSE images were 0.54 mm vs 0.53 mm and 1.96 mm vs 1.68 mm,respectively;and the control points of the large distortions were distributed at the edges of the phantom,which was consistent with the MR imaging characteristics previously reported.Compared with CT-based segmentation contour,the MDA was 1.17 mm and DSC was 0.91 for 3D T1WI-FFE,while MDA was 0.86 mm and DSC was 0.94 for 3D T2WI-TSE.Conclusion The study quantitatively assesses the geometric accuracy of the imaging system for MR-guided radiotherapy.The phantom-based contour analysis reveals that with CT image as gold standard,the segmentation error in MRI images meets the clinical requirements,and that 3D T2WI-TSE image is advantageous over 3D T1WI-FFE image in segmentation accuracy.

Objective Magnetic resonance simulator(MR Sim)is a novel type of simulation equipment utilized in radiotherapy.Acceptance testing is an essential quality assurance procedure prior to the clinical use of the MR Sim.The report provides the detailed procedures and result analysis of acceptance testing for an MR Sim.Methods The acceptance testing scheme was developed following the recently published AAPM TG284 report and the NCC/T-RT 002-2023 guidelines.Quality control equipments such as ACR(American College of Radiology)large phantom and geometric distortion measurement phantom were used for evaluating various aspects of the MR Sim,including the effectiveness of shielding,the functionality of imaging system,the image quality,the performance of radio frequency coils,the geometric accuracy of large field imaging,the precision of external laser markings,the couch movement accuracy,and the image transmission accuracy.Results The shielding effectiveness at a frequency of 150 MHz exhibited an average value of 105 dB.All of 8 image quality indices,namely geometric accuracy,slice position accuracy,slice thickness accuracy,image uniformity,artifact ratio,signal-to-noise ratio,high-contrast spatial resolution,and low-contrast resolution,fell within recommended tolerances.The maximum geometric distortion observed across a 25 cm field of view was less than 2 mm.The errors in external laser markings and couch movement accuracy were both less than 1 mm.The couch levelness was less than 1°.Both radio frequency coils and image transmission passed the required tests.Conclusion MR Sim is high-precision and complex.To ensure its precise application in radiotherapy,the acceptance testing for an MR Sim should be meticulously designed and executed following the established guidelines and accounting for its unique performance characteristics.

Objective:To evaluate the feasibility of using MRI-only simulation images for dose calculation of both photon and proton radiotherapy for nasopharyngeal carcinoma cases.Methods:T 1-weighted MRI images and CT images of 100 patients with nasopharyngeal carcinoma treated with radiotherapy in Cancer Hospital of Chinese Academy of Medical Sciences from January 2020 to December 2021 were retrospectively analyzed. MRI images were converted to generate pseudo-CT images by using deep learning network models. The training set, validation set and test set included 70 cases, 10 cases and 20 cases, respectively. Convolutional neural network (CNN) and cycle-consistent generative adversarial neural network (CycleGAN) were exploited. Quantitative assessment of image quality was conducted by using mean absolute error (MAE) and structural similarity (SSIM), etc. Dose assessment was performed by using 3D-gamma pass rate and dose-volume histogram (DVH). The quality of pseudo-CT images generated was statistically analyzed by Wilcoxon signed-rank test. Results:The MAE of the CNN and CycleGAN was (91.99±19.98) HU and (108.30±20.54) HU, and the SSIM was 0.97±0.01 and 0.96±0.01, respectively. In terms of dosimetry, the accuracy of pseudo-CT for photon dose calculation was higher than that of the proton plan. For CNN, the gamma pass rate (3 mm/3%) of the photon radiotherapy plan was 99.90%±0.13%. For CycleGAN, the value was 99.87%±0.34%. The gamma pass rates of proton radiotherapy plans were 98.65%±0.64% (CNN, 3 mm/3%) and 97.69%±0.86% (CycleGAN, 3 mm/3%). For DVH, the dose calculation accuracy in the photon plan of pseudo-CT was better than that of the proton plan.Conclusions:The deep learning-based model generated accurate pseudo-CT images from MR images. Most dosimetric differences were within clinically acceptable criteria for photon and proton radiotherapy, demonstrating the feasibility of an MRI-only workflow for radiotherapy of nasopharyngeal cancer. However, compared with the raw CT images, the error of the CT value in the nasal cavity of the pseudo-CT images was relatively large and special attention should be paid during clinical application.