Objective:An integrated automatic quality control system(automatic quality control instrument)for high dose rate(HDR)afterloader was independently developed.This system was used to measure the placement accuracy,residence time and activity of the radioactive source,so as to assess the accuracy and stability of this system in quality assurance.Methods:The automatic quality control system was connected to the Flexitron type of HDR afterloader through a connecting tube.Quality assurance(QA)plans with different resident positions,clearances,and times were designed for measurement and verification.The measurement length was 1079-1119 mm,and the measurement time was 2-10 s,and the measurement activity was 14,326-28,653 μGy·m2·h-1.The measurements were repeated once per one week in three months,so as to determine the consistency of the measurement parameters in a longer period of time.The measurement results of the placement accuracy,resident time and activity of radioactive source of the HDR 192Ir radioactive source were analyzed.Results:The plate ionization chamber was used to measure the activity of radioactive source.In three months,the measurement deviation of the activity of radioactive source was 1.17%,and the fitted activity decay curve was consistent with the expected decay curve of radioactive source.The measurement accuracies of the positioning of radioactive source were respectively 0.46,0.18,0.23,0.25,and 0.15 mm when the interval step sizes of resident times were respectively 2,3,5,8,and 10 mm,all of them were<1 mm.The measurement accuracy of time was(4.90±0.04)s when the resident time was 5 s.In the measurement accuracy of different resident times(1,2,3,5 and 10s)of same step size,the deviation of them was less than 0.5s.When the resident time was the shortest,the deviation between the read-out value of the automatic system software of quality control and the actually resident time was(0.09±0.04)s.The measurement accuracy gradually stabilized with the increasing of resident times.Conclusion:The automatic quality control instrument can complete the measurements for the positioning accuracy,the accuracy of resident time and the activity of radioactive source,and achieve rapidly and accurately periodic quality assurance.

Objective:To apply artificial intelligence（AI）in classifying ovarian tumors on ultrasound images，and compare the diagnostic results of several sonographers with varying seniority levels.Methods:A total of 645 patients diagnosed with adnexal masses via gynecological ultrasound examination at Qilu Hospital of Shandong University from January 2021 to December 2024 were enrolled. Three deep learning architectures，i.e.，Alexnet，Densenet121，and Resnet50 were developed and used to internally test the classification effectiveness of ovarian tumors，while the optimal model was selected for external testing. Two junior sonographers and two senior sonographers were recruited to independently diagnose ovarian tumors in the external test dataset. Subsequently，the benign and malignant results of the model's predictions were disclosed to each sonographer，and their revised diagnoses on the same external test data in combination with the best AI model were recorded.Results:The optimal model achieved an accuracy of 0.941，sensitivity of 0.936，and specificity of 0.944 on the internal test dataset，and maintained robust performance on the external test dataset with accuracy of 0.891，sensitivity of 0.880，and specificity of 0.907. Compared to junior sonographers，the optimal model demonstrated significantly higher sensitivity in discriminating benign from malignant ovarian tumors（0.880 vs. 0.723，0.602；all P<0.05）. No statistically significant difference was observed in diagnostic accuracy between the optimal model and senior sonographer 1（ P=0.05）. With assistance from the optimal model，junior sonographers achieved significant improvements in both sensitivity and specificity（sensitivity：0.723 vs. 0.843，0.602 vs. 0.819；specificity：0.778 vs. 0.833，0.685 vs. 0.741；all P<0.05）. Conclusions:The optimal model achieves comparable performance to that of senior sonographers in ovarian tumor classification. With model assistance，the diagnostic performance of junior sonographers is significantly improved.

Objective:An integrated automatic quality control system(automatic quality control instrument)for high dose rate(HDR)afterloader was independently developed.This system was used to measure the placement accuracy,residence time and activity of the radioactive source,so as to assess the accuracy and stability of this system in quality assurance.Methods:The automatic quality control system was connected to the Flexitron type of HDR afterloader through a connecting tube.Quality assurance(QA)plans with different resident positions,clearances,and times were designed for measurement and verification.The measurement length was 1079-1119 mm,and the measurement time was 2-10 s,and the measurement activity was 14,326-28,653 μGy·m2·h-1.The measurements were repeated once per one week in three months,so as to determine the consistency of the measurement parameters in a longer period of time.The measurement results of the placement accuracy,resident time and activity of radioactive source of the HDR 192Ir radioactive source were analyzed.Results:The plate ionization chamber was used to measure the activity of radioactive source.In three months,the measurement deviation of the activity of radioactive source was 1.17%,and the fitted activity decay curve was consistent with the expected decay curve of radioactive source.The measurement accuracies of the positioning of radioactive source were respectively 0.46,0.18,0.23,0.25,and 0.15 mm when the interval step sizes of resident times were respectively 2,3,5,8,and 10 mm,all of them were<1 mm.The measurement accuracy of time was(4.90±0.04)s when the resident time was 5 s.In the measurement accuracy of different resident times(1,2,3,5 and 10s)of same step size,the deviation of them was less than 0.5s.When the resident time was the shortest,the deviation between the read-out value of the automatic system software of quality control and the actually resident time was(0.09±0.04)s.The measurement accuracy gradually stabilized with the increasing of resident times.Conclusion:The automatic quality control instrument can complete the measurements for the positioning accuracy,the accuracy of resident time and the activity of radioactive source,and achieve rapidly and accurately periodic quality assurance.

Objective:To apply artificial intelligence（AI）in classifying ovarian tumors on ultrasound images，and compare the diagnostic results of several sonographers with varying seniority levels.Methods:A total of 645 patients diagnosed with adnexal masses via gynecological ultrasound examination at Qilu Hospital of Shandong University from January 2021 to December 2024 were enrolled. Three deep learning architectures，i.e.，Alexnet，Densenet121，and Resnet50 were developed and used to internally test the classification effectiveness of ovarian tumors，while the optimal model was selected for external testing. Two junior sonographers and two senior sonographers were recruited to independently diagnose ovarian tumors in the external test dataset. Subsequently，the benign and malignant results of the model's predictions were disclosed to each sonographer，and their revised diagnoses on the same external test data in combination with the best AI model were recorded.Results:The optimal model achieved an accuracy of 0.941，sensitivity of 0.936，and specificity of 0.944 on the internal test dataset，and maintained robust performance on the external test dataset with accuracy of 0.891，sensitivity of 0.880，and specificity of 0.907. Compared to junior sonographers，the optimal model demonstrated significantly higher sensitivity in discriminating benign from malignant ovarian tumors（0.880 vs. 0.723，0.602；all P<0.05）. No statistically significant difference was observed in diagnostic accuracy between the optimal model and senior sonographer 1（ P=0.05）. With assistance from the optimal model，junior sonographers achieved significant improvements in both sensitivity and specificity（sensitivity：0.723 vs. 0.843，0.602 vs. 0.819；specificity：0.778 vs. 0.833，0.685 vs. 0.741；all P<0.05）. Conclusions:The optimal model achieves comparable performance to that of senior sonographers in ovarian tumor classification. With model assistance，the diagnostic performance of junior sonographers is significantly improved.

Using big data and artificial intelligence to establish a multi-point monitoring, early warning, and disposal system to achieve early warning and intervention of infectious disease outbreaks is an important means of controlling the spread of the epidemic. Taking Xiaoshan district as an example, this study analyzes the monitoring contents, warning methods, and application effectiveness of the infectious disease monitoring, early warning and disposal system. Based on Xiaoshan′s health big data resources, the system starts with syndrome, disease diagnosis and etiology. Through advanced technologies such as artificial intelligence and block chain, it realizes early identification of infectious disease outbreaks, data fusion, multi-cross collaboration, and closed-loop management. It has improved the sensitivity of clustered outbreaks monitoring and the effectiveness of epidemic disposal and provided a reference for grassroots disease prevention and control departments to establish an infectious disease monitoring and early warning system.

Using big data and artificial intelligence to establish a multi-point monitoring, early warning, and disposal system to achieve early warning and intervention of infectious disease outbreaks is an important means of controlling the spread of the epidemic. Taking Xiaoshan district as an example, this study analyzes the monitoring contents, warning methods, and application effectiveness of the infectious disease monitoring, early warning and disposal system. Based on Xiaoshan′s health big data resources, the system starts with syndrome, disease diagnosis and etiology. Through advanced technologies such as artificial intelligence and block chain, it realizes early identification of infectious disease outbreaks, data fusion, multi-cross collaboration, and closed-loop management. It has improved the sensitivity of clustered outbreaks monitoring and the effectiveness of epidemic disposal and provided a reference for grassroots disease prevention and control departments to establish an infectious disease monitoring and early warning system.

Objective:To investigate the dosimetry effect of rotational errors of multi-channel cylinder vaginal applicator of intravaginal irradiation after surgery of endometrial cancer.Methods:A total of 18 patients who underwent surgery of endometrial cancer at Peking Union Medical College Hospital from June to December 2022 were selected.The plans of patients who adopted the treatment of multi-channel cylinder applicator of vagina were retrospectively analyzed,which maintained the same retained mode with clinical plan.The applicator was rotated clockwise by 22.5? and 45.0?,respectively,simulating the rotational errors that occurred in placing the applicator among clinical inter-fractions.And then,the changes of dosimetry of target area and organs at risk(OAR)under two kinds of rotation amplitudes were further analyzed.Results:When the applicator was rotated as 22.5?,the minimum doses to 90%volumes of CTV by 2.03%than that of clinical plan,which was significantly different(t=5.86,P<0.05),and the maximal doses to 2cc of OARs of bladder and rectum respectively increased 2.35%and 2.71%,and the differences of them were statistically significant(t=-3.49,-2.40,P<0.05),respectively.When the applicator was rotated as 45?,the D90 of the target area decreased by 5.75%than that of clinical plan,which was statistically significant(t=14.07,P<0.05).The D2cc values of the bladder and rectum increased respectively by 6.50%and 9.49%than that of clinical plan,which were statistically significant(t=-7.72,-6.9,P<0.05).The differences of the exposed doses of sigmoid colon and small intestine after the applicator was rotated by 22.5? and 45.0? between the plan and original plan were respectively less,which were not statistical significance.Conclusion:The multi-channel cylinder applicator can provide individualized dose distribution in intravaginal irradiation.However,attention should be paid to the placement of the applicator when patients undergo inter-fractional treatment,in order to avoid deviations in the angular alignment from the original plan.This can impact the dosages of the target area and OARs.

Purpose: Mixed-lineage leukemia protein 4 (MLL4/KMT2D) is a histone methyltransferase, and its mutation has been reported to be associated with a poor prognosis in many cancers, including lung cancer. We investigated the function of MLL4 in lung carcinogenesis. Materials and Methods: RNA sequencing (RNA-seq) in A549 cells transfected with control siRNA or MLL4 siRNA was performed. Also, we used EdU incorporation assay, colony formation assays, growth curve analysis, transwell invasion assays, immunohistochemical staining, and in vivo bioluminescence assay to investigate the function of MLL4 in lung carcinogenesis. Results: We found that MLL4 expression was downregulated in non–small cell lung cancer (NSCLC) tissues compared to adjacent normal tissues and tended to decrease with disease stage progression. We analyzed the transcriptomes in control and MLL4- deficient cells using high-throughput RNA deep sequencing (RNA-seq) and identified a cohort of target genes, such as SOX2, ATF1, FOXP4, PIK3IP1, SIRT4, TENT5B, and LFNG, some of which are related to proliferation and metastasis. Our results showed that low expression of MLL4 promotes NSCLC cell proliferation and metastasis and is required for the maintenance of NSCLC stem cell properties. Conclusion Our findings identify an important role of MLL4 in lung carcinogenesis through transcriptional regulation of PIK3IP1, affecting the PI3K/AKT/SOX2 axis, and suggest that MLL4 could be a potential prognostic indicator and target for NSCLC therapy.

Objective:To formulate clinical commissioning procedures, items, and testing method for Flexitron afterloader hardware and software and establish relevant quality control procedures in order to meet national standards and clinical requirements.Methods:Clinical commissioning included hardware, treatment planning system (TPS), and end-to-end (ETE) full-process testing. The radioactive source positioning accuracy was measured using a source position check ruler. The accuracy and linearity of dwell time were evaluated using three method: stopwatch timing, ion chamber measurement, and video analysis. The accuracy of source position simulator, connecting tubes, source position check ruler, and other measuring tools was tested using a high-precision ruler. Films were used for calibration of marker lines and applicators. The electrometer and well chamber were used to calibrate the radioactive source activity. The display and reconstruction accuracies of the TPS were evaluated using physical image data. ETE testing was conducted using a custom-made phantom for scanning, planning, and dose measurement.Results:The accuracy testing result of the commissioning items were within acceptable limits. The deviation in source activity measurements was 0.21%, and the ETE point dose measurement deviation was 2.32%, both of which met the clinical requirements. However, there was a 2 mm difference between the nominal and measured values of the magnetic resonance marker line in the accuracy testing items. Therefore, adjustments were required when using marker line for catheter reconstruction based on magnetic resonance images.Conclusions:By summarizing the clinical commissioning experience of the Flexitron afterloader, this study has developed quality control method and baseline levels of result of afterloader and TPS items, which provides a reference for the commissioning before clinical use.

Brachytherapy (BT) provides the opportunity to deliver highly potent radiation doses to the tumor, while more effectively sparing the surrounding organs at risk (OAR) due to the proximity of radiation sources to the tumor target and rapid fall-off of the source dose profile. As an important part of radiotherapy for cervical cancer, BT plays an irreplaceable role. The BT process is complex and can be divided into a series of steps. Long time waiting for patients in the state of implantation of the applicator may cause changes in the position of the applicator relative to the tumor and the movement of the OAR. In recent years, artificial intelligence (AI) has made great progress in the medical field. Machine learning and neural network have been widely applied in all aspects of BT, such as implantation of the applicator, image acquisition, segmentation of target area and OAR, reconstruction of the applicator, plan optimization, and treatment delivery, etc. In addition, BT significantly reduces the overall time, improves the homogeneity of operation, and enhances the accuracy of treatment. In this article, the application, development prospects, and challenges of AI in the BT of cervical cancer in recent years were reviewed, aiming to provide novel ideas for the application of AI in BT.