Purpose This study aimed to evaluate the consistency of human epidermal growth factor receptor 2(HER2)status between primary breast cancer lesions and lung metastatic lesions and to establish a prognostic model for predicting the survival rate of HER2 low expression(HER2-low)breast cancer patients with lung metastasis.Methods Clinicopathological data from a cohort of 252 patients with breast cancer and lung metastasis were retrospec-tively analyzed.Results 50.00％of the patients had HER2-low expression in metastatic lesions,and HER2-low ex-pression was the most prevalent subgroup in both primary and metastatic lesions.A discordance in HER2 status be-tween primary and metastatic sites was observed in 28.07％of cases.The most frequent shift was from HER2-zero in the primary tumor to HER2-low expression in the metastasis(12.28％of all cases).Estrogen receptor(ER)status,menopausal status,and histological type were identified as independent prognostic factors for overall survival(OS)by univariate and multivariate Cox regression analyses.A prognostic model incorporating these factors was constructed to predict 3-year and 5-year survival.The model demonstrated area under the curve(AUC)values of 0.765 and 0.780 for 3-year and 5-year OS in the training cohort,and 0.667 and 0.706 in the validation cohort,respectively.Conclu-sion HER2-low expression is the most common subtype among breast cancer patients with lung metastasis.The ob-served shift from HER2-zero in primary lesions to HER2-low in metastases underscores the clinical necessity of re-biop-sy at metastatic sites.The developed prognostic model effectively predicts OS in this patient population.

Objective To develop a three-dimensional(3D)deep neural network based preoperative classification model for coronary computed tomography angiography(CTA)in atrial fibrillation patients,and to explore the effects of transfer learning on the performance of medical image classification models,thereby providing preoperative decision support for catheter ablation to advance atrial fibrillation treatment toward precision and personalization.Methods Utilizing 3D ConvNet and 3D ResNet as backbone network,the three-dimensional classification features were extracted from coronary CTA sequences.The publicly available pre-trained weights were used for transfer learning.The model performance was evaluated through metrics such as confusion matrix,classification accuracy,and area under the curve(AUC).A comparative analysis was also conducted to evaluate the performance differences between the transfer learning model and the initialized training model.Results Transfer learning yielded significant performance improvements over the initialized training models,attaining AUC improvement of 9.1%-16.7%and accuracy enhancement of 6.2%-23.5%.Among all models,3D-ResNet18 model with MedicalNet pre-training weights performed the best,achieving an AUC of 0.77 and an accuracy of 0.71.Conclusion The proposed three-dimensional deep network enhanced by transfer learning can effectively identify atrial fibrillation patients requiring additional ablation besides pulmonary vein isolation through preoperative coronary CTA,which will assist clinicians in optimizing surgical strategies and improving treatment outcomes,thereby reducing long-term postoperative recurrence rates.

Purpose This study aimed to evaluate the consistency of human epidermal growth factor receptor 2(HER2)status between primary breast cancer lesions and lung metastatic lesions and to establish a prognostic model for predicting the survival rate of HER2 low expression(HER2-low)breast cancer patients with lung metastasis.Methods Clinicopathological data from a cohort of 252 patients with breast cancer and lung metastasis were retrospec-tively analyzed.Results 50.00％of the patients had HER2-low expression in metastatic lesions,and HER2-low ex-pression was the most prevalent subgroup in both primary and metastatic lesions.A discordance in HER2 status be-tween primary and metastatic sites was observed in 28.07％of cases.The most frequent shift was from HER2-zero in the primary tumor to HER2-low expression in the metastasis(12.28％of all cases).Estrogen receptor(ER)status,menopausal status,and histological type were identified as independent prognostic factors for overall survival(OS)by univariate and multivariate Cox regression analyses.A prognostic model incorporating these factors was constructed to predict 3-year and 5-year survival.The model demonstrated area under the curve(AUC)values of 0.765 and 0.780 for 3-year and 5-year OS in the training cohort,and 0.667 and 0.706 in the validation cohort,respectively.Conclu-sion HER2-low expression is the most common subtype among breast cancer patients with lung metastasis.The ob-served shift from HER2-zero in primary lesions to HER2-low in metastases underscores the clinical necessity of re-biop-sy at metastatic sites.The developed prognostic model effectively predicts OS in this patient population.

Objective To develop a three-dimensional(3D)deep neural network based preoperative classification model for coronary computed tomography angiography(CTA)in atrial fibrillation patients,and to explore the effects of transfer learning on the performance of medical image classification models,thereby providing preoperative decision support for catheter ablation to advance atrial fibrillation treatment toward precision and personalization.Methods Utilizing 3D ConvNet and 3D ResNet as backbone network,the three-dimensional classification features were extracted from coronary CTA sequences.The publicly available pre-trained weights were used for transfer learning.The model performance was evaluated through metrics such as confusion matrix,classification accuracy,and area under the curve(AUC).A comparative analysis was also conducted to evaluate the performance differences between the transfer learning model and the initialized training model.Results Transfer learning yielded significant performance improvements over the initialized training models,attaining AUC improvement of 9.1%-16.7%and accuracy enhancement of 6.2%-23.5%.Among all models,3D-ResNet18 model with MedicalNet pre-training weights performed the best,achieving an AUC of 0.77 and an accuracy of 0.71.Conclusion The proposed three-dimensional deep network enhanced by transfer learning can effectively identify atrial fibrillation patients requiring additional ablation besides pulmonary vein isolation through preoperative coronary CTA,which will assist clinicians in optimizing surgical strategies and improving treatment outcomes,thereby reducing long-term postoperative recurrence rates.

Objective:To analyze the short-term therapeutic efficacy of radial extracorporeal shock wave therapy for patients with subacromial impingement syndrome.Methods:A total of 106 patients diagnosed as having subacromial impingement syndrome between October 2017 and April 2019 were randomized into a radial extracorporeal shock wave therapy (rESWT) group of 36, an exercise rehabilitation group of 35 and a conventional therapy group of 35. In addition to family exercise therapy, the rESWT group underwent 2000 to 2500 shots of extracorporeal shock wave therapy at 10 Hz and a pressure of 1.5-2.5 bar, once a week for four consecutive weeks. The exercise group was given range of motion exercises, joint control training and tendon movement training for 45 minutes, three times a week for four consecutive weeks. The conventional therapy group was treated with a laser apparatus and low-frequency electrotherapy, once a day, three times a week. Constant-Murle scores (CMSs) and the short form health survey (SF-36) were used to evaluate the clinical efficacy before and after 1 month of treatment.Results:Before the treatment there were no significant differences among the 3 groups in any of the measurements. After one month of treatment the average CMS pain score and total score of the exercise rehabilitation group were significantly better than the conventional therapy group′s averages. Moreover, the average body pain score, daily life ability, range of motion, muscle strength and total score of the rESWT group were all significantly better than the exercise and conventional therapy groups′ averages. In the SF-36 the average physical function, bodily pain, general health, and mental health scores of the rESWT groups were also significantly better than the other 2 groups′ averages.Conclusions:Radial extracorporeal shock wave therapy is superior to exercise therapy and conventional therapy for patients with subacromial impingement syndrome. It can restore shoulder joint function and improve the quality of life in one month.

Objective To observe the effect of surface electromyographic biofeedback (sEMG BFB) combined with routine swallow training in treating dysphagia among those with nasopharyngeal carcinoma after radiation therapy.Methods Fifty dysphagic patients with nasopharyngeal carcinoma after radiation therapy were randomly divided into a biofeedback training group and a routine treatment group,each of 25.Both groups were given routine training including orofacial function training,sensory irritation,behavioral swallowing training,and electrical stimulation.The biofeedback group was additionally given behavioral swallowing training based on sEMG BFB.Before and 4 weeks after the treatment,a videofluoroscopic swallowing study was performed to observe the opening of the upper esophageal sphincter (UES).The penetration aspiration scale (PAS) and the functional oral intake scale (FOIS) were used to evaluate the subjects' swallowing function.Results Before the treatment there were no significant differences between the two groups in terms of UES opening,average PAS score or average FOIS score.Everyone improved significantly after the treatment,but compared with the routine treatment group,UES opening was significantly better after the treatment,the average PAS score was lower and the average FOIS score was higher in the biofeedback training group.Conclusion sEMG BFB combined with routine swallowing training can improve the UES opening and swallowing ability of dysphagic patients with nasopharyngeal carcinoma after radiation therapy.

Objective To explore the factor structure of the anxiety imagery characteristics questionnaire.Methods Through the literature review and expert group discussion,the initial anxiety imagery features questionnaire was created.The first questionnaire contained 49 items of anxiety imagery,which was answered by 106 imagery dialogue psychotherapists.After analysis of the items and exploratory factors,the formal questionnaire that contained 20 items of anxiety imagery characteristics was formed.The formal questionnaire was measured by 115 imagery dialogue psychotherapists,then,exploratory and confirmatory factor analyses were made.Results The exploratory factor analysis showed that four factors whose Eigen value were more than 1 were extracted from l0 items,including emotional panic,uneasy,tense muscles,and motor restlessness.Those four factors explained 54.39％ of the total variance.Confirmatory analysis showed that x2/df was 2.142 ; root mean square error of approximation (RMSEA) was 0.059 ; and comparative fit index (CFI),goodness-of-fit index (GFI),non-normed fit index (NNFI) and RSMEA was 0.912,0.935,0.896,and 0.057,respectively.Conclusions The four-factor structure was constructed by 10 items of anxiety imagery characteristics questionnaire.