Objective:To construct a machine learning classification prediction model using planning-omics (planomics) features to predict the γ pass rate of intensity-modulated radiotherapy (IMRT) plan dose verification in fixed-field thoracic tumors, and evaluate the application of planomics in radiotherapy quality assurance.Methods:The fixed-field IMRT plans of 240 patients with chest tumors admitted to Department of Radiotherapy, Henan Cancer Hospital from August 2022 to March 2023 were retrospectively analyzed. All plans underwent dose verification using the electronic portal imaging system detector on the Varian accelerator to collect field dose data. The dose verification results were analyzed through Portal Dosimetry in the treatment planning system of Eclipse. The γ pass rate standard was set at 2%/2 mm with a 10% dose threshold. From the planning documents, 48 conventional planning features, 2476 planomics features, and the combination of the previous two feature sets were extracted. Subsequently, an auto-encoder classification model was constructed. To evaluate the classification efficacy of various feature sets, 20 random train-test divisions were conducted by calculating the area under the receiver operating characteristic curve (AUC) values along with the accuracy rates.Results:After the feature selection, 2 conventional features and 16 planomics features were finally selected. In the testing set, the AUC values for the model using combined features, planomics features, and conventional planned features were 0.802±0.030, 0.740±0.069, and 0.673±0.083, respectively. In contrast, in the training set, these AUC values were 0.844±0.074, 0.816±0.047, and 0.687±0.036, respectively. The accuracy rates were 0.752±0.083, 0.703±0.110, and 0.648±0.081 in the testing set, and 0.753±0.098, 0.751±0.075, and 0.624±0.054 in the training set for the combined, planomics, and conventional planning feature sets, respectively.Conclusions:For thoracic fixed-field adjusted radiotherapy planning, the machine learning method based on planomics features can be utilized to build a classification model for predicting the γ pass rate. Combining planomics features with conventional planned features can enhance the predictive performance of the classification models.

Objective:To evaluate the detection capability of the contrast-enhanced three-dimensional turbo spin-echo (CE-SPACE) sequence for brain metastases (BM), aiming to provide evidence for precise target delineation in stereotactic radiotherapy (SRT).Methods:A total of 123 BM patients who received radiotherapy at the Affiliated Cancer Hospital of Zhengzhou University from May to November 2024 were enrolled. All patients underwent contrast-enhanced (CE) MRI and CT scans in the same treatment position, with images rigidly registered in the Eclipse planning system. Two experienced radiation oncologists independently delineated BM lesions on CE-MPRAGE and CE-SPACE sequences in a blinded manner. Patients were divided into the delayed group (10 min, n=61) and a priority group (5 min, n=62) based on the time interval between gadolinium injection and CE-SPACE acquisition. The non-parametric Wilcoxon rank-sum test was used to compare the lesion counts and volume differences between the two imaging sequences. Point-biserial correlation analysis was performed to assess the correlation between the additional lesions identified by CE-SPACE and lesion volume. Results:The overall analysis demonstrated that CE-SPACE detected 421 BM lesions, achieving an 8.2% higher detection rate than CE-MPRAGE ( Z=3.78, P<0.001). In terms of lesion volume, the median BM lesions volume identified by CE-SPACE [0.30(0.07,1.53)cm 3] was 8.7% larger than that by CE-MPRAGE [0.23 (0.04, 1.34) cm 3] ( Z=12.88, P<0.001). CE-SPACE demonstrated superior sensitivity for lesions ≤ 0.06 cm3, with negative correlation between the number of additional lesions detected and lesion volume ( r=-0.104, P=0.034). Subgroup analysis revealed that in the delayed group, CE-SPACE detected significantly more lesions [median 2 (1, 3.5) vs. 2 (1, 3), P=0.002] and larger volumes [0.39 (0.08, 2.24) cm3 vs. 0.29 (0.05, 1.99) cm3, P<0.001] than CE-MPRAGE. In the priority group, CE-SPACE detected significantly larger lesion volumes [0.55 (0.09, 2.06) cm3 vs. 0.45 (0.08, 1.88) cm3, P<0.001], but no significant difference was observed in lesion counts between two sequences ( P=0.059). Conclusions:Three-dimensional CE-SPACE sequence offers superior detection sensitivity for small BM (≤ 0.06 cm3), providing crucial guidance for accurate target delineation in SRT.

Objective To explore the predictive efficacy of several multimodal MRI-based machine learning models for the promoter methylation states of O6-methylguanine-DNA methyltransferase(MGMT)of glioblastoma muliforme(GBM)patients in terms of the GBM heterogeneity and the complexity of the tumor microenvironment.Methods Firstly,the multimodal MRI images of 317 GBM patients from The University of Pennsylvania Glioblastoma(UPENN-GBM)dataset were pre-processed,with four sequences involved in including T1-weighted imaging(T1WI)sequence,T1-weighted contrast-enhanced imaging(T1CE)sequence,T2-weighted imaging(T2WI)sequence and fluid-attenuated inversion recovery(FLAIR)sequence,and the radiomics features were extracted for two regions of interest(ROIs)such as the tumor core region and the tumor edema region.Secondly,the data of the 317 GBM patients were randomly divided into a training set(254 cases)and a test set(63 cases),which underwent normalization with Z-scores and feature selection and dimensionality reduction with Lasso regression.Finally,three models were established respectively with particle swarm optimization-support vector machine(PSO-SVM),C-support vector classification(C-SVC)and adaptive boosting(adaptive boosting(Adaboost)algorithms,and the predictive efficacy of the three models for glioblastoma multiforme MGMT promoter methylation states were evaluated in terms of accuracy and AUC.Results The Adaboost model based on T2WI sequence and radiomics features of the tumor core region had the highest predictive efficacy with accuracy and AUC values of 67％and 0.74,respectively,higher than those of other combinations of sequences,models and regions of interest.Conclusion The multimodal MRI-based machine learning models can be used for the prediction of glioblastoma multiforme MGMT promoter methylation states,which provides powerful support for personalized treatment and prognostic assessment of GBM.[Chinese Medical Equipment Journal,2025,46(6):7-13]

Objective To investigate the oral health status of second-grade primary school students in Suzhou,provide baseline data for evaluating Suzhou's"Bright Eyes and Healthy Teeth"public health project,and offer reference for formulating oral health policies for students in Suzhou.Methods A multi-stage,stratified,random sampling method was used to select 1 640 second-grade primary school students from 10 districts(county-level cities)across Suzhou.The oral health survey of primary school students in Suzhou was designedbased on WHO"Basic Methods for Oral Health Surveys"(5th edition).A survey on the students' oral health status was con-ducted.Results The caries prevalence,mean DMFT(Decayed,Missing,Filled Teeth),and caries filling rate for primary teeth among second-grade students in Suzhou were 80.73％,4.25,and 24.48％,respectively.For permanent teeth,the caries prevalence,mean DMFT(Decayed,Missing,Filled Teeth),and caries filling rate were 30.30％,0.62,and 16.96％,respectively.Specifically,the caries prevalence,mean DMFT,and filling rate for the first permanent molar were 29.88％,0.59,and 16.94％,respectively.The pit and fissure sealing rate for the first permanent molar was 13.74％.Additionally,the detection rate for gingival bleeding was 58.05％,and for dental calculus,49.15％.Conclusion The oral health status of second-grade primary school students in Suzhou is concerning.The caries prevalence rate for both primary and permanent teeth is high,while the filling rate is relatively low.Prior to the implementation of oral health programs,the pit and fissure sealing rate for the first permanent molars was low,with significant differ-ences observed between urban and rural areas,as well as among different districts(or county-level cities).

Objective In order to provide a better reference and basis for the selection of reasonable hypoglycemic drugs for clinical treatment,the study conducted a comprehensive clinical evaluation of the innovator sodium-glucose transporters 2(SGLT-2)inhibitors,based on A Quick Guideline for Drug Evaluation and Selection in Chinese Medical Institutions(the Second Edition).Methods The real-world studies,randomized controlled trials,Meta-analysis/systematic review,drug clinical use guidelines,expert consensus and drug description evaluation evidence were collected,and the included drugs were assigned and evaluated from five dimensions:pharmaceutical characteristics,efficacy,safety,economy and other attributes.Results All SGLT-2 inhibitors had evaluation scores above 75,with dagaglifloztin tablets having the highest score of 84.6,and canaglifloztin having the lowest score of 75.1.Conclusions All five original SGLT-2 inhibitors showed good clinical utility,the difference is that the participating original drugs have different advantageous intervals in clinical use.The results show that dagliflozin has the most ideal clinical utility,and its clinical use should be safer and more effective.Due to the short time on the market and insufficient evidence-based reasons,the advantages of clinical use of proline hemegliflozin are not obvious compared with other evaluated drugs.

Objective To investigate the oral health status of second-grade primary school students in Suzhou,provide baseline data for evaluating Suzhou's"Bright Eyes and Healthy Teeth"public health project,and offer reference for formulating oral health policies for students in Suzhou.Methods A multi-stage,stratified,random sampling method was used to select 1 640 second-grade primary school students from 10 districts(county-level cities)across Suzhou.The oral health survey of primary school students in Suzhou was designedbased on WHO"Basic Methods for Oral Health Surveys"(5th edition).A survey on the students' oral health status was con-ducted.Results The caries prevalence,mean DMFT(Decayed,Missing,Filled Teeth),and caries filling rate for primary teeth among second-grade students in Suzhou were 80.73％,4.25,and 24.48％,respectively.For permanent teeth,the caries prevalence,mean DMFT(Decayed,Missing,Filled Teeth),and caries filling rate were 30.30％,0.62,and 16.96％,respectively.Specifically,the caries prevalence,mean DMFT,and filling rate for the first permanent molar were 29.88％,0.59,and 16.94％,respectively.The pit and fissure sealing rate for the first permanent molar was 13.74％.Additionally,the detection rate for gingival bleeding was 58.05％,and for dental calculus,49.15％.Conclusion The oral health status of second-grade primary school students in Suzhou is concerning.The caries prevalence rate for both primary and permanent teeth is high,while the filling rate is relatively low.Prior to the implementation of oral health programs,the pit and fissure sealing rate for the first permanent molars was low,with significant differ-ences observed between urban and rural areas,as well as among different districts(or county-level cities).

Objective To explore the predictive efficacy of several multimodal MRI-based machine learning models for the promoter methylation states of O6-methylguanine-DNA methyltransferase(MGMT)of glioblastoma muliforme(GBM)patients in terms of the GBM heterogeneity and the complexity of the tumor microenvironment.Methods Firstly,the multimodal MRI images of 317 GBM patients from The University of Pennsylvania Glioblastoma(UPENN-GBM)dataset were pre-processed,with four sequences involved in including T1-weighted imaging(T1WI)sequence,T1-weighted contrast-enhanced imaging(T1CE)sequence,T2-weighted imaging(T2WI)sequence and fluid-attenuated inversion recovery(FLAIR)sequence,and the radiomics features were extracted for two regions of interest(ROIs)such as the tumor core region and the tumor edema region.Secondly,the data of the 317 GBM patients were randomly divided into a training set(254 cases)and a test set(63 cases),which underwent normalization with Z-scores and feature selection and dimensionality reduction with Lasso regression.Finally,three models were established respectively with particle swarm optimization-support vector machine(PSO-SVM),C-support vector classification(C-SVC)and adaptive boosting(adaptive boosting(Adaboost)algorithms,and the predictive efficacy of the three models for glioblastoma multiforme MGMT promoter methylation states were evaluated in terms of accuracy and AUC.Results The Adaboost model based on T2WI sequence and radiomics features of the tumor core region had the highest predictive efficacy with accuracy and AUC values of 67％and 0.74,respectively,higher than those of other combinations of sequences,models and regions of interest.Conclusion The multimodal MRI-based machine learning models can be used for the prediction of glioblastoma multiforme MGMT promoter methylation states,which provides powerful support for personalized treatment and prognostic assessment of GBM.[Chinese Medical Equipment Journal,2025,46(6):7-13]

ObjectiveTo investigate the epidemiological characteristics of human parainfluenza virus (HPIV) in Pudong New Area, Shanghai, and to provide a basis for the prevention and control of HPIV infections in this area. MethodsA total of 8 180 cases with acute respiratory infection (ARI)/influenza-like illness (ILI) attending the fever outpatient clinics, pediatric outpatient clinics, respiratory outpatient clinics, pediatric wards, emergency departments, respiratory wards, and intensive care units (ICUs) and other monitoring departments in 9 sentinel hospitals in Pudong New Area from 2017 to 2022 were selected as the research subjects, and their nasopharyngeal/ throat swabs were collected. Polymerase chain reaction (PCR) method was performed for testing the respiratory virus such as influenza virus, human adenovirus, human parainfluenza virus, respiratory syncytial virus (RSV), human enterovirus, rhinovirus, et al. HPIV serotypes were analyzed, and the detection rates of HPIV were compared between different times, seasons, and population groups. ResultsThe overall HPIV detection rate of ARI/ILI cases in Pudong New Area was 3.73% (305/8 180) in 2017‒2022, with HPIV-3 being the predominant serotype. The detection rate of HPIV decreased significantly in 2017‒2022, peaking at 7.66% (99/1 293) in 2017 and falling to 0.80% (13/1 617) in 2021. Statistically significant differences were observed in HPIV detection rates across different years (χ²_trend=80.037, P_trend<0.001). The detection rate was higher in 2017‒2019 than that in 2020‒2022 (χ²_trend=38.990, P_trend<0.001). HPIV exhibited seasonal patterns in 2017‒2019, with higher detection rates in summer and autumn and lower in spring and winter, whereas no seasonal patterns were observed in 2020‒2022. Children aged <6 years had the highest detection rate (7.07%, 139/1 967), followed by adults aged ≥60 years (4.78%, 85/1 779). Statistically significant differences were observed in the detection rates of HPIV among different age groups (χ²=111.210, P<0.001). Symptoms of HPIV infection were predominantly cough, fever, and runny nose in pre-school children, fever, cough, and sore throat in school-age children, and adults, and cough, expectoration, and fever in the elderly. ConclusionThe HPIV detection rate in Pudong New Area was low in 2017‒2022. The seasonal pattern of HPIV circulation in Pudong New Area disappeared in 2020‒2022 due to the influence of infectious disease epidemics. Young children and the elderly should be prioritized in HPIV prevention and control efforts.

Objective To investigate the effect and mechanism of circular RNA_0009910(circ_0009910)on the proliferation and apoptosis of non-small cell lung cancer cells.Methods A549 cells were randomly divided into groups and transfected as follows:si-NC group,si-circ_0009910 group,miR-NC group,miR-34a-5p group,si-circ_0009910+anti-miR-NC group,and si-circ_0009910+anti-miR-34a-5p group.After successful transfec-tion,cell proliferation was measured using the CCK-8 assay.Dual-luciferase reporter assays were used to vali-date the target gene of circ_0009910.Apoptosis of A549 cells was analyzed by flow cytometry.Protein expres-sion levels of B-cell lymphoma-2(Bcl-2),transmembrane receptor protein 1(Notch1),cell proliferation-asso-ciated nuclear antigen(Ki-67),Bcl-2-associated X protein(Bax),and cysteine aspartic acid protease-3(Caspase-3)were detected by Western blot.Cells from each group were subcutaneously injected into the abdo-mens of male nude mice to establish xenograft models.Tumor volume,tumor weight,and protein expression in tumor tissues were measured.Results Compared with the si-NC group,the si-circ_0009910 group showed significantly reduced A549 cell proliferation,increased apoptosis,decreased expression of Ki-67,Notch1,and Bcl-2,and increased expression of Bax and Caspase-3(P＜0.05).Compared with the miR-NC group,the miR-34a-5p group exhibited significantly reduced luciferase activity after transfection with wild-type circ_0009910(P＜0.05).The si-circ_0009910 group showed upregulated miR-34a-5p expression compared to the si-NC group.The miR-34a-5p group demonstrated weaker proliferation,higher apoptosis,downregulated Bcl-2,Notch1,and Ki-67,and upregulated Bax and Caspase-3 compared to the miR-NC group(P＜0.05).Compared with the si-circ_0009910+anti-miR-NC group,the si-circ_0009910+anti-miR-34a-5p group exhibited en-hanced proliferation,reduced apoptosis,upregulated Bcl-2,Notch1,and Ki-67,and downregulated Bax and Caspase-3(P＜0.05).In vivo,the si-circ_0009910 and miR-34a-5p groups showed smaller tumor volumes and weights,downregulated Notch1,Ki-67,and Bcl-2,and upregulated Bax and Caspase-3 in tumor tissues com-pared to their respective control groups(P＜0.05).Conclusion Low expression of circ_0009910 inhibits A549 cell proliferation and promotes apoptosis,potentially via the miR-34a-5p/Notch1 signaling pathway.

Objective To investigate the characteristics of thickness changes in peripapillary retinal nerve fiber layer(pRNFL)and identify related risk factors in patients with Moyamoya disease(MMD).Methods A retrospective study was conducted on 150 MMD patients(150 eyes)aged 6-65 years admitted to the Neurosurgery Department of the Fifth Medical Center,Chinese PLA General Hospital from May 2016 to December 2023(observation group),and 150 age-matched healthy volunteers(150 eyes)from the hospital's ophthalmology outpatient department(control group).Both groups were subdivided into pediatric(≤18 years),young adult(18-40 years),and middle-aged(40-65 years)subgroups.The pRNFL thickness in four quadrants was measured by optical coherence tomography(OCT):superior(pRNFL-Sup),inferior(pRNFL-Inf),nasal(pRNFL-Nas),temporal(pRNFL-Tmp),and average thickness(pRNFL-Avg).General clinical data and pRNFL thickness were compared between two groups.Univariate and multivariate logistic regression analyses were performed to identify risk factors for pRNFL thinning in MMD patients.The cohort was randomly divided into training(n=210)and validation(n=90)sets at a 7:3 ratio.A predictive model for pRNFL thinning in MMD patients was constructed based on logistic regression results.Model performance was evaluated using the area under the receiver operating characteristic curve(AUC),and clinical utility was assessed via decision curve analysis.Results Compared with control group,MMD patients exhibited significantly reduced pRNFL-Avg,pRNFL-Sup,pRNFL-Tmp,and pRNFL-Inf thickness(P<0.05 or P<0.001),while pRNFL-Nas showed no significant difference(P>0.05).In the pediatric subgroup,pRNFL-Avg and pRNFL-Inf were thinner(P<0.05).In the young adult subgroup,pRNFL-Avg and pRNFL-Sup were reduced(P<0.001 or P<0.05).In the middle-aged subgroup,pRNFL-Avg,pRNFL-Sup,pRNFL-Inf,and pRNFL-Tmp were all thinner(P<0.05 or P<0.001).Multivariate logistic regression identified visual field defects(OR=15.28,95%CI 2.95-79.10),disease duration(OR=1.11,95%CI 1.05-1.18),and the number of involved cerebral vessels(OR=1.49,95%CI 1.01-2.22)as independent risk factors for pRNFL thinning.The predictive model achieved AUC of 0.94(95%CI 0.91-0.97)and 0.95(95%CI 0.91-0.99)in the training and validation sets,respectively.Decision curve analysis confirmed the model's favorable clinical net benefit.Conclusion Thinning of pRNFL was observed in Moyamoya disease patients with visual field defects,disease duration,and cerebral vascular involvement identified as independent risk factors for pRNFL atrophy.