1.The dual roles of NCompass nitinol stone retrieval basket in ureteroscopic holmium laser lithotripsy for upper ureteral stones larger than 1 cm
Xing CHEN ; Yangtian JIAO ; Yuhui HE ; Yisen DENG ; Mingxiao ZHANG ; Yiwei TAN ; Jianfeng WANG
Journal of Modern Urology 2026;31(1):17-20
Objective To analyze the application of NCompass nitinol stone retrieval basket in ureteroscopic holmium laser lithotripsy, and to evaluate its roles in reducing stone retropulsion and improving stone-free rate(SFR)and stone acquisition rate.Methods A retrospective analysis was conducted on the data of patients with upper ureteral stones>1 cm who underwent ureteroscopic holmium laser lithotripsy at China-Japan Friendship Hospital during Jan.2018 and Dec.2023, including 96 cases were in the NCompass group, and 68 cases in the non-basket group.The operation time, stone retropulsion rate, usage rate of flexible ureteroscopy, SFR and stone acquisition rate were compared between the two groups.Results There were no significant differences in the operation time [(29.2±8.4)min vs.(30.9±7.2)min, P=0.170] and stone retropulsion rate(15.6% vs. 27.9%, P=0.055)between the NCompass group and non-basket group.The NCompass group had significantly lower usage rate of flexible ureteroscopy than the non-basket group(24.0% vs. 47.1%, P=0.002), higher SFR 2 days(80.2% vs. 45.6%, P<0.001)and 1 month(93.8% vs. 83.8%, P=0.040)after operation, as well as the stone acquisition rates 2 days(100% vs. 44.1%, P<0.001)and 1 month(100% vs. 58.8%, P<0.001)after operation. Conclusion The NCompass nitinol stone retrieval basket plays dual roles in ureteroscopic holmium laser lithotripsy, which can effectively intercept stones and extract stone fragments, thus significantly improving the SFR and stone acquisition rate, and reducing stone retropulsion and the use of flexible ureteroscopy.
2.Expert recommendations on the development content and functional specifications for the public vaccination service platform
Qi ZHU ; Qianli MA ; Ruili XIE ; Lijun LIU ; Lei LI ; Lin CHEN ; Yong HUANG ; Ronghai TAN ; Xiaoru CAI ; Jianfeng HE ; Wenzhou YU
Chinese Journal of Preventive Medicine 2025;59(9):1448-1453
To satisfy the growing healthcare demands of the public, it is essential to develop a public service platform for vaccination. This initiative aligns with national policies, optimizes resource allocation, innovates service models, enhances service efficiency, and reduces service costs. Drawing on relevant national policies and regulatory requirements, as well as the notable achievements and practical experiences gained through the exploration and innovation of vaccination service models across various regions, this paper proposes expert recommendations. It defines the essential components and functional specifications for public service platforms, focusing on public needs such as electronic vaccination record management, appointment management, the promotion of electronic vaccination certificates, vaccination certificate verification for school enrollment, vaccination site navigation, and science communication and public engagement. The recommendations aim to serve as a reference for the development of vaccination public service platforms nationwide.
3.Dosimetric study of radiotherapy synchronized with 3D printing-based tumor treating fields for glioblastoma
Zhongwei LI ; Xuwei LU ; Di WU ; Jianfeng TAN ; Zaijie HUANG ; Pei YANG ; Yujuan ZHOU ; Hong LIU
Chinese Journal of Medical Physics 2025;42(6):712-718
Objective To investigate the dosimetric effects of tumor treating fields(TTFields)patches on different radiotherapy modes for glioblastoma(GBM)patients who wear TTFields patches during radiotherapy,thereby providing dosimetric guidance for determining the appropriate radiotherapy mode.Methods With the TTFields data from GBM patients,artifact-free radiotherapy CT images were obtained utilizing 3D-printed TPU TTFields patches(3D-Print-TTFields)and anthropomorphic phantoms,and then a TTFields-synchronized radiotherapy image model was constructed.Furthermore,the treatment planning system was used to construct a dosimetric calculation model for TTFields-synchronized radiotherapy by simulating and fitting the ray attenuation rate of TTFields patches measured by accelerators.Using these models,3 kinds of radiotherapy plans were simulated and developed.Specifically,P1 simulated the conventional radiotherapy mode;P2 simulated the TTFields-combined radiotherapy mode(TTF-Com-RT),in which patients underwent radiotherapy using the P1 plan while wearing TTFields patches;and P3 simulated the TTFields-synchronized radiotherapy(TTF-Syn-RT)mode where the TTFields patches were worn throughout the entire radiotherapy process.The paired t-test was used to analyze dosimetric parameters such as target dose(D95),average scalp dose(D-skin),conformity index(CI)and homogeneity index(HI)in 3 plans(P1,P2,and P3),as well as the D95 and D-skin parameters for intensity-modulated radiotherapy(IMRT)and volumetric modulated arc therapy(VMAT)techniques in the P3 plan.Results The D95 simulated by P2 decreased by 1.35%as compared with P1(P<0.05),and the D95 simulated by P3 was 1.31%higher than that in P2(P<0.05).Compared with P1,P2 and P3 increased the D-skin by 12.56%and 14.30%,respectively(P<0.05),and the D-skin simulated by P3 increased by 1.55%as compared with P2(P<0.05).However,there were trivial differences in D95 between P3 and P1,CI and HI among all plans,D95 and D-skin between IMRT and VMAT techniques in P3 plan(P>0.05).Conclusion Based on GBM patient data,CT simulation images obtained from 3D-Print-TTFields combined with anthropomorphic phantom are artifact-free and meet radiotherapy requirements.The target and scalp dose differences between TTF-Com-RT and TTF-Syn-RT are less than 2%,and the dosimetric difference of TTF-Syn-RT using IMRT/VMAT techniques is insignificant.Therefore,clinicians can choose radiotherapy modes and techniques according to actual needs.
4.Expert recommendations on the development content and functional specifications for the public vaccination service platform
Qi ZHU ; Qianli MA ; Ruili XIE ; Lijun LIU ; Lei LI ; Lin CHEN ; Yong HUANG ; Ronghai TAN ; Xiaoru CAI ; Jianfeng HE ; Wenzhou YU
Chinese Journal of Preventive Medicine 2025;59(9):1448-1453
To satisfy the growing healthcare demands of the public, it is essential to develop a public service platform for vaccination. This initiative aligns with national policies, optimizes resource allocation, innovates service models, enhances service efficiency, and reduces service costs. Drawing on relevant national policies and regulatory requirements, as well as the notable achievements and practical experiences gained through the exploration and innovation of vaccination service models across various regions, this paper proposes expert recommendations. It defines the essential components and functional specifications for public service platforms, focusing on public needs such as electronic vaccination record management, appointment management, the promotion of electronic vaccination certificates, vaccination certificate verification for school enrollment, vaccination site navigation, and science communication and public engagement. The recommendations aim to serve as a reference for the development of vaccination public service platforms nationwide.
5.Influence of emergency endovascular treatment on the prognosis of minor stroke caused by posterior circulation large vessel occlusion
Xi CHEN ; Chen CHEN ; Yufei GENG ; Suhang SHANG ; Wenfeng SONG ; Suixia CAO ; Ying TAN ; Jia YU ; Jianfeng HAN
Chinese Journal of Cerebrovascular Diseases 2025;22(10):677-689
Objective To investigate the effects of emergency endovascular treatment on the short-term and long-term prognosis of patients with minor stroke(National Institutes of Health stroke scale[NIHSS]score≤5)caused by posterior circulation large vessel occlusion(LVO).Methods A retrospective analysis was performed on consecutive patients with minor stroke caused by posterior circulation LVO admitted to the Department of Neurology,the First Affiliated Hospital of Xi'an Jiaotong University from July 2019 to March 2024.The patients were divided into the emergency endovascular treatment group and the standard medical treatment group according to the treatment method.Baseline and clinical data were collected from all patients enrolled,including age,sex,smoking history,history of alcohol consumption,medical history(hypertension,diabetes,hyperlipidemia,atrial fibrillation,transient ischemic attack[TIA],blood pressure on admission,stroke history,coronary heart disease),intravenous thrombolysis,tandem lesions,posterior circulation Alberta stroke program early CT score(pc-ASPECTS)on admission,NIHSS score on admission and discharge,time from onset to admission,responsible occluded vessel(basilar artery,left vertebral artery,right vertebral artery),vertebral artery development(left vertebral artery dominant,right vertebral artery dominant,bilateral vertebral artery dominant),non-lesion side vertebral artery development(poor,good,not applicable),basilar artery on CT angiography(BATMAN)score,leptomeningeal branch compensation(open,not open),surgery-related indicators(number of thrombectomy passes[≤2 times,>2 times],rescue interventions[stent placement,balloon dilation,arterial thrombolysis,intra-arterial tirofiban infusion],immediate postoperative modified thrombolysis in cerebral infarction[mTICI]grade≥2b[successful recanalization],anesthesia method[general,local],endotracheal intubation status[yes,no],duration of mechanical ventilation[not using a ventilator or successfully intubation for≤24 hours and>24 hours]),in-hospital systematic complications(deep-vein thrombosis,urinary tract infection,lung infection).The primary outcome for short-term prognosis was an excellent outcome(modified Rankin scale[mRS]score of 0-1)within 90 days after onset.Secondary outcomes included a good outcome(mRS score of 0-2)within 90days after onset,recurrent ischemic stroke within 90 days after onset,all-cause mortality within 90 days after onset.Safety outcomes were symptomatic intracerebral hemorrhage(sICH)within 24 hours of treatment(NIHSS score increased by≥4 points or increased level of consciousness score by≥1 point compared with admission,with visible hemorrhagic lesions on follow-up CT scan)and early neurological deterioration(END,NIHSS score increased by≥2 points or motor score increased by≥1 point compared with admission,within 24 hours after treatment).Long-term outcome was defined as recurrent ischemic stroke within 1 year after onset.Short-term and safety outcomes were compared between the emergency endovascular treatment group and the standard medical treatment group.Kaplan-Meier survival curves was used to evaluate the effect of emergency endovascular treatment on the long-term prognosis.Based on the mRS score at 90 days from onset,all patients were divided into an excellent outcome(mRS score 0-1)group and a non-excellent outcome(mRS score 2-6)group.Baseline and clinical data were compared across the two groups.Variables with statistically significant differences were included in the multivariate Logistic regression analysis to investigate the influencing factor of 90-day excellent outcomes in patients with minor stroke caused by posterior circulation LVO.Results A total of 56 patients with minor stroke caused by posterior circulation LVO were enrolled,including 18 patients in the emergency endovascular treatment group and 38 patients in the standard medical treatment group.45 patients achieved excellent outcomes and 11 patients achieved non-excellent outcomes.(1)The emergency endovascular treatment group had lower pc-ASPECTS on admission(8.0[7.0,9.0]points vs.9.0[8.0,10.0]points,P=0.043)and There were no statistically significant differences in the excellent outcome rate,good outcome rate,and ischemic stroke recurrence rate within 90 days after onset between the two groups(all P>0.05).No all-cause mortality occurred within 90 days after onset in either group.In the emergency endovascular treatment group,one patient developed sICH and one developed END within 24 hours after treatment.(3)No recurrent ischemic stroke in the emergency endovascular treatment group within 1 year after onset,while 3cases(7.89%)of recurrence were observed within 1year after onset in the standard medical treatment group.The Kaplan-Meier survival curve analysis showed that there was no statistically significant difference in the incidence of ischemic stroke within one year after onset between the two groups(P=0.341).(4)There were statistically significant differences between patients with excellent outcome and patients with non excellent outcome in drinking history,diabetes history,NIHSS score after discharge,distribution of responsible occlusive vessels,and distribution of vertebral artery development(all P<0.05).The results of multivariate Logistic regression analysis showed that the NIHSS score at discharge was an independent influencing factor for excellent outcome at 90 days after onset in patients with minor stroke caused by posterior circulation LVO(OR,0.448,95%CI 0.275-0.728,P=0.001).Conclusions This study shows potential safety and effectiveness of emergency endovascular treatment on patients with minor stroke caused by posterior circulation LVO,but it is not superior to standard medical treatment in terms of short-term and long-term outcomes.Further large-sample randomized controlled trials are warranted to validate the findings of this study.
6.Influence of emergency endovascular treatment on the prognosis of minor stroke caused by posterior circulation large vessel occlusion
Xi CHEN ; Chen CHEN ; Yufei GENG ; Suhang SHANG ; Wenfeng SONG ; Suixia CAO ; Ying TAN ; Jia YU ; Jianfeng HAN
Chinese Journal of Cerebrovascular Diseases 2025;22(10):677-689
Objective To investigate the effects of emergency endovascular treatment on the short-term and long-term prognosis of patients with minor stroke(National Institutes of Health stroke scale[NIHSS]score≤5)caused by posterior circulation large vessel occlusion(LVO).Methods A retrospective analysis was performed on consecutive patients with minor stroke caused by posterior circulation LVO admitted to the Department of Neurology,the First Affiliated Hospital of Xi'an Jiaotong University from July 2019 to March 2024.The patients were divided into the emergency endovascular treatment group and the standard medical treatment group according to the treatment method.Baseline and clinical data were collected from all patients enrolled,including age,sex,smoking history,history of alcohol consumption,medical history(hypertension,diabetes,hyperlipidemia,atrial fibrillation,transient ischemic attack[TIA],blood pressure on admission,stroke history,coronary heart disease),intravenous thrombolysis,tandem lesions,posterior circulation Alberta stroke program early CT score(pc-ASPECTS)on admission,NIHSS score on admission and discharge,time from onset to admission,responsible occluded vessel(basilar artery,left vertebral artery,right vertebral artery),vertebral artery development(left vertebral artery dominant,right vertebral artery dominant,bilateral vertebral artery dominant),non-lesion side vertebral artery development(poor,good,not applicable),basilar artery on CT angiography(BATMAN)score,leptomeningeal branch compensation(open,not open),surgery-related indicators(number of thrombectomy passes[≤2 times,>2 times],rescue interventions[stent placement,balloon dilation,arterial thrombolysis,intra-arterial tirofiban infusion],immediate postoperative modified thrombolysis in cerebral infarction[mTICI]grade≥2b[successful recanalization],anesthesia method[general,local],endotracheal intubation status[yes,no],duration of mechanical ventilation[not using a ventilator or successfully intubation for≤24 hours and>24 hours]),in-hospital systematic complications(deep-vein thrombosis,urinary tract infection,lung infection).The primary outcome for short-term prognosis was an excellent outcome(modified Rankin scale[mRS]score of 0-1)within 90 days after onset.Secondary outcomes included a good outcome(mRS score of 0-2)within 90days after onset,recurrent ischemic stroke within 90 days after onset,all-cause mortality within 90 days after onset.Safety outcomes were symptomatic intracerebral hemorrhage(sICH)within 24 hours of treatment(NIHSS score increased by≥4 points or increased level of consciousness score by≥1 point compared with admission,with visible hemorrhagic lesions on follow-up CT scan)and early neurological deterioration(END,NIHSS score increased by≥2 points or motor score increased by≥1 point compared with admission,within 24 hours after treatment).Long-term outcome was defined as recurrent ischemic stroke within 1 year after onset.Short-term and safety outcomes were compared between the emergency endovascular treatment group and the standard medical treatment group.Kaplan-Meier survival curves was used to evaluate the effect of emergency endovascular treatment on the long-term prognosis.Based on the mRS score at 90 days from onset,all patients were divided into an excellent outcome(mRS score 0-1)group and a non-excellent outcome(mRS score 2-6)group.Baseline and clinical data were compared across the two groups.Variables with statistically significant differences were included in the multivariate Logistic regression analysis to investigate the influencing factor of 90-day excellent outcomes in patients with minor stroke caused by posterior circulation LVO.Results A total of 56 patients with minor stroke caused by posterior circulation LVO were enrolled,including 18 patients in the emergency endovascular treatment group and 38 patients in the standard medical treatment group.45 patients achieved excellent outcomes and 11 patients achieved non-excellent outcomes.(1)The emergency endovascular treatment group had lower pc-ASPECTS on admission(8.0[7.0,9.0]points vs.9.0[8.0,10.0]points,P=0.043)and There were no statistically significant differences in the excellent outcome rate,good outcome rate,and ischemic stroke recurrence rate within 90 days after onset between the two groups(all P>0.05).No all-cause mortality occurred within 90 days after onset in either group.In the emergency endovascular treatment group,one patient developed sICH and one developed END within 24 hours after treatment.(3)No recurrent ischemic stroke in the emergency endovascular treatment group within 1 year after onset,while 3cases(7.89%)of recurrence were observed within 1year after onset in the standard medical treatment group.The Kaplan-Meier survival curve analysis showed that there was no statistically significant difference in the incidence of ischemic stroke within one year after onset between the two groups(P=0.341).(4)There were statistically significant differences between patients with excellent outcome and patients with non excellent outcome in drinking history,diabetes history,NIHSS score after discharge,distribution of responsible occlusive vessels,and distribution of vertebral artery development(all P<0.05).The results of multivariate Logistic regression analysis showed that the NIHSS score at discharge was an independent influencing factor for excellent outcome at 90 days after onset in patients with minor stroke caused by posterior circulation LVO(OR,0.448,95%CI 0.275-0.728,P=0.001).Conclusions This study shows potential safety and effectiveness of emergency endovascular treatment on patients with minor stroke caused by posterior circulation LVO,but it is not superior to standard medical treatment in terms of short-term and long-term outcomes.Further large-sample randomized controlled trials are warranted to validate the findings of this study.
7.Dosimetric study of radiotherapy synchronized with 3D printing-based tumor treating fields for glioblastoma
Zhongwei LI ; Xuwei LU ; Di WU ; Jianfeng TAN ; Zaijie HUANG ; Pei YANG ; Yujuan ZHOU ; Hong LIU
Chinese Journal of Medical Physics 2025;42(6):712-718
Objective To investigate the dosimetric effects of tumor treating fields(TTFields)patches on different radiotherapy modes for glioblastoma(GBM)patients who wear TTFields patches during radiotherapy,thereby providing dosimetric guidance for determining the appropriate radiotherapy mode.Methods With the TTFields data from GBM patients,artifact-free radiotherapy CT images were obtained utilizing 3D-printed TPU TTFields patches(3D-Print-TTFields)and anthropomorphic phantoms,and then a TTFields-synchronized radiotherapy image model was constructed.Furthermore,the treatment planning system was used to construct a dosimetric calculation model for TTFields-synchronized radiotherapy by simulating and fitting the ray attenuation rate of TTFields patches measured by accelerators.Using these models,3 kinds of radiotherapy plans were simulated and developed.Specifically,P1 simulated the conventional radiotherapy mode;P2 simulated the TTFields-combined radiotherapy mode(TTF-Com-RT),in which patients underwent radiotherapy using the P1 plan while wearing TTFields patches;and P3 simulated the TTFields-synchronized radiotherapy(TTF-Syn-RT)mode where the TTFields patches were worn throughout the entire radiotherapy process.The paired t-test was used to analyze dosimetric parameters such as target dose(D95),average scalp dose(D-skin),conformity index(CI)and homogeneity index(HI)in 3 plans(P1,P2,and P3),as well as the D95 and D-skin parameters for intensity-modulated radiotherapy(IMRT)and volumetric modulated arc therapy(VMAT)techniques in the P3 plan.Results The D95 simulated by P2 decreased by 1.35%as compared with P1(P<0.05),and the D95 simulated by P3 was 1.31%higher than that in P2(P<0.05).Compared with P1,P2 and P3 increased the D-skin by 12.56%and 14.30%,respectively(P<0.05),and the D-skin simulated by P3 increased by 1.55%as compared with P2(P<0.05).However,there were trivial differences in D95 between P3 and P1,CI and HI among all plans,D95 and D-skin between IMRT and VMAT techniques in P3 plan(P>0.05).Conclusion Based on GBM patient data,CT simulation images obtained from 3D-Print-TTFields combined with anthropomorphic phantom are artifact-free and meet radiotherapy requirements.The target and scalp dose differences between TTF-Com-RT and TTF-Syn-RT are less than 2%,and the dosimetric difference of TTF-Syn-RT using IMRT/VMAT techniques is insignificant.Therefore,clinicians can choose radiotherapy modes and techniques according to actual needs.
8.Effect of low-dose esketamine for postoperative analgesia on postoperative depression in patients with gastrointestinal tumors
Jie GAO ; Xi ZHANG ; Yuanyuan RONG ; Tao HU ; Yan GAO ; Bibo TAN ; Jianfeng FU ; Huaqin LIU
Chinese Journal of Anesthesiology 2024;44(7):797-801
Objective:To evaluate the effect of low-dose esketamine for postoperative analgesia on the postoperative depression in patients with gastrointestinal tumors.Methods:This study was a prospective randomized controlled trial. Eighty patients, aged 18-64 yr, with a body mass index of 18-25 kg/m 2, of American Society of Anesthesiologists Physical Status classification Ⅱ or Ⅲ, scheduled for elective radical resection of the gastrointestinal tumor under general anesthesia from June to November 2023 in our hospital, were divided into 2 groups ( n=40 each) using a random number table method: esketamine group (group E) and control group (group C). Each patient received postoperative patient-controlled intravenous analgesia(PCIA). The PCIA solution in group E contained esketamine 0.5 mg/kg, dezocine 0.5 mg/kg, dexmetomidine 1.5 μg/kg and flurbiprofen ester 100 mg in 100 ml of normal saline. The PCIA solution in group C contained dezocine 0.5 mg/kg, dexmetomidine 1.5 μg/kg and flurbiprofen ester 100 mg in 100 ml of normal saline. The Hospital Anxiety and Depression Scale (HADS) was used to assess the patients′ anxiety and depression at 1 day before operation (T 0) and 2 days after operation (T 1). The Quality of Recovery-15 scale was used to evaluate the early postoperative recovery quality. Visual analog scale scores, the pressing times of patient-controlled analgesia and the number of rescue analgesia were recorded within 2 days after operation. The occurrence of drug-related adverse reactions was also recorded. Results:Seventy-eight patients were finally included, with 39 cases in group E and 39 cases in group C. Compared with group C, the postoperative HADS-depression scale score and incidence of depression were significantly decreased, the Quality of Recovery-15 scale score was increased, the visual analog scale scores were decreased ( P<0.05), and no significant changes were found in the postoperative HADS-anxiety scale score and incidence of anxiety, the pressing times of patient-controlled analgesia and the number of rescue analgesia in group E ( P>0.05). Visual hallucination was found at 1 day after operation in one patient and relieved at 2 days after operation in group E. There was no significant difference in the incidence of postoperative dizziness, nausea and vomiting between the two groups ( P>0.05). Conclusions:Postoperative analgesia with 0.5 mg/kg esketamine can alleviate postoperative depressive symptoms, enhance the efficacy of analgesia and improve the early postoperative recovery quality in patients with gastrointestinal tumors.
9.The principle and practice of vidian neurectomy
Changqing ZHAO ; Xicai SUN ; Yuzhu WAN ; Jing YE ; Guolin TAN ; Jianfeng LIU ; Yanjie WANG ; Fengli CHENG ; Yunfang AN
Chinese Journal of Otorhinolaryngology Head and Neck Surgery 2024;59(1):51-56
The latest research findings on bidirectional regulation of neuro-immunity through traditional neural circuits shed new light on the theoretical basis of the role of vidian neurectomy (VN). This article aims to provide a comprehensive understanding of VN, including the history of VN, the principle of neuroimmuno-interaction, the applied anatomy of VN as well as the methods of transnasal endoscopic surgery. Additionally, we introduce the concept of the nose-brain axis, which was proposed based on the advancement in the area of neuro-immune interactions.
10.A multicenter study on the prediction of gamma passing rate based on radiomic features
Luqiao CHEN ; Qianxi NI ; Yu WU ; Huan REN ; Jinmeng PANG ; Jianfeng TAN ; Longjun LUO ; Zhili WU ; Jinjia CAO
Chinese Journal of Radiological Medicine and Protection 2024;44(12):1027-1033
Objective:To construct classification prediction models for gamma passing rate using radiomics-based machine learning approaches and data from multiple radiotherapy institutions and evaluate the models′ performance.Methods:The data from 572 volumetric-modulated arc therapy (VMAT) patients across three radiotherapy institutions (514 for training and 58 for testing)were retrospectively collected. Additionally, 45 VMAT plans were collected from a single institution as an independent external validation set. For all the data, a three-dimensional dose validation approach based on actual measurements of phantoms was utilized, and gamma analysis was performed at the 3%/2 mm criterion using a dose threshold of 10%, absolute doses, and global normalization. After radiomic features were extracted from dose files, feature selection was performed using the random forest (RF) method and RF combined with Shapley Additive exPlanation (SHAP). Then, feature subsets of varying sizes (10, 20, 30, 40, and 50) were selected based on feature rankings. Using these subsets as inputs, data training was conducted using the Extreme Gradient Boosting (XGBoost) algorithm. Finally, the models′ classification performance was assessed using the area under the curve (AUC) values and F1-score.Results:Under the 3%/2 mm criterion, all models performed the best in the case of 20 feature subsets. The optimal prediction model established based on the feature selection using RF exhibited AUC and F1-score of 0.88 and 0.89, respectively on the testing set and 0.82 and 0.90, respectively, on the validation set. The optimal prediction model built based on the feature selection using RF combined with SHAP yielded AUC and F1-score of 0.86 and 0.92 on the testing set and 0.87 and 0.89, respectively, on the validation set, along with superior robustness. Therefore, the second model possessed certain advantages over the first model.Conclusions:For multicenter dose verification result, it is feasible to construct a machine learning prediction model with high classification performance using radiomic features derived from dose files, combined with feature selection based on SHAP. This approach can assist in advancing the clinical applications and implementation of gamma passing rate prediction models.

Result Analysis
Print
Save
E-mail