Objective:To develop a deep learning model based on super-resolution endorectal ultrasound（ERUS）images for the preoperative prediction of perineural invasion（PNI）in patients with rectal cancer，thereby providing a reference for risk stratification and individualized treatment planning.Methods:A retrospective analysis was conducted on 382 patients with rectal cancer who underwent total mesorectal excision at Fujian Medical University Union Hospital between June 2019 and February 2024. Patients were randomly divided into a training set（ n=305）and a test set（ n=77）at a ratio of 8∶2，and further grouped into PNI-negative group and PNI-positive group subgroups based on pathological results. Super-resolution ultrasound images were generated from original ERUS images using a generative adversarial network（GAN）. Deep convolutional neural networks were developed based on features from intratumoral and peritumoral regions to identify the optimal region of interest（ROI）. The dSR5_ResNet18 and dSR5_ResNet50 models were constructed using the super-resolution images with a 5-pixel peritumoral extension. Representative clinical features were selected for subgroup analysis based on sample size and intergroup statistical differences between PNI-positive and PNI-negative patients. Forest plots were used to evaluate model applicability and robustness across subgroups. Results:The dSR5_ResNet18 model，built using super-resolution images of the tumor combined with a 5-pixel peritumoral region，achieved the best predictive performance，with an AUC of 0.867（95% CI=0.782 - 0.952）in the test set. Decision curve analysis demonstrated that the dSR5_ResNet18 model provided the greatest net clinical benefit. Forest plot analysis indicated strong generalizability of the models across subgroups such as pathological N stage，maximum lesion length，and lymph node enlargement，though relatively weaker performance was observed in the carcinoembryonic antigen（CEA）subgroup. Among all models，dSR5_ResNet18 exhibited the most consistent performance across subgroups，with the narrowest confidence intervals and highest robustness. Conclusions:The deep learning model incorporating ERUS-based super-resolution reconstruction demonstrated excellent performance in the preoperative prediction of PNI in rectal cancer. It offers significant advantages in image quality and generalizability，and may serve as a valuable tool to assist clinicians in formulating personalized treatment strategies.

Objective:To investigate the implementation status of periodic risk evaluation reports(PRER)for medical devices in China,so as to provide references for adapting the development of vigilance system for medical device,and improving the PRER system.Methods:A questionnaire was self-designed,which was online distributed to registrants and filing persons for medical devices,domestic agent for imported medical devices,and the relevant persons of monitoring agencies for adverse events of medical device at all grades through monitoring agencies of each provincial grade in China,so as to understand the evaluation and suggestion of the relevant agencies of medical devices for the PRER work.Combined with review status and special inspection results of all monitoring agencies for PRER,and compared with guidelines of the European Union's Periodic Safety Update Reports(PSURs),the existing issues in the PRER for medical devices in China were analyzed at present.Results:A total of 4,377 valid questionnaires were returned,the respondents by 64.95%acknowledged that PRER can aid stakeholders to identify products'risks.Additionally,the respondents by 41.28%acknowledged that the medical devices of Class I and renewal registrations need be simplify in the PRER,and the respondents by 57.69%considered the frequency of writing and submitting reports should be adjusted.Conclusion:This article combined with the requirements of pilot work of vigilance for medical device,and recommended to strengthen risk analysis in PRER,and streamline reporting formats,and improve the top-level design of safety supervision after medical devices enter the market in China.These measures will promote the supervision and management system of China for medical device to integrate with international standards,and enhance the international competitiveness of the domestically medical device industry.

Objective:To investigate the implementation status of periodic risk evaluation reports(PRER)for medical devices in China,so as to provide references for adapting the development of vigilance system for medical device,and improving the PRER system.Methods:A questionnaire was self-designed,which was online distributed to registrants and filing persons for medical devices,domestic agent for imported medical devices,and the relevant persons of monitoring agencies for adverse events of medical device at all grades through monitoring agencies of each provincial grade in China,so as to understand the evaluation and suggestion of the relevant agencies of medical devices for the PRER work.Combined with review status and special inspection results of all monitoring agencies for PRER,and compared with guidelines of the European Union's Periodic Safety Update Reports(PSURs),the existing issues in the PRER for medical devices in China were analyzed at present.Results:A total of 4,377 valid questionnaires were returned,the respondents by 64.95%acknowledged that PRER can aid stakeholders to identify products'risks.Additionally,the respondents by 41.28%acknowledged that the medical devices of Class I and renewal registrations need be simplify in the PRER,and the respondents by 57.69%considered the frequency of writing and submitting reports should be adjusted.Conclusion:This article combined with the requirements of pilot work of vigilance for medical device,and recommended to strengthen risk analysis in PRER,and streamline reporting formats,and improve the top-level design of safety supervision after medical devices enter the market in China.These measures will promote the supervision and management system of China for medical device to integrate with international standards,and enhance the international competitiveness of the domestically medical device industry.

Objective:To develop a deep learning model based on super-resolution endorectal ultrasound（ERUS）images for the preoperative prediction of perineural invasion（PNI）in patients with rectal cancer，thereby providing a reference for risk stratification and individualized treatment planning.Methods:A retrospective analysis was conducted on 382 patients with rectal cancer who underwent total mesorectal excision at Fujian Medical University Union Hospital between June 2019 and February 2024. Patients were randomly divided into a training set（ n=305）and a test set（ n=77）at a ratio of 8∶2，and further grouped into PNI-negative group and PNI-positive group subgroups based on pathological results. Super-resolution ultrasound images were generated from original ERUS images using a generative adversarial network（GAN）. Deep convolutional neural networks were developed based on features from intratumoral and peritumoral regions to identify the optimal region of interest（ROI）. The dSR5_ResNet18 and dSR5_ResNet50 models were constructed using the super-resolution images with a 5-pixel peritumoral extension. Representative clinical features were selected for subgroup analysis based on sample size and intergroup statistical differences between PNI-positive and PNI-negative patients. Forest plots were used to evaluate model applicability and robustness across subgroups. Results:The dSR5_ResNet18 model，built using super-resolution images of the tumor combined with a 5-pixel peritumoral region，achieved the best predictive performance，with an AUC of 0.867（95% CI=0.782 - 0.952）in the test set. Decision curve analysis demonstrated that the dSR5_ResNet18 model provided the greatest net clinical benefit. Forest plot analysis indicated strong generalizability of the models across subgroups such as pathological N stage，maximum lesion length，and lymph node enlargement，though relatively weaker performance was observed in the carcinoembryonic antigen（CEA）subgroup. Among all models，dSR5_ResNet18 exhibited the most consistent performance across subgroups，with the narrowest confidence intervals and highest robustness. Conclusions:The deep learning model incorporating ERUS-based super-resolution reconstruction demonstrated excellent performance in the preoperative prediction of PNI in rectal cancer. It offers significant advantages in image quality and generalizability，and may serve as a valuable tool to assist clinicians in formulating personalized treatment strategies.

ObjectiveTo evaluate the clinical efficacy and safety of Tongluo Mingmu capsules in the treatment of diabetic retinopathy with blood stasis， collateral obstruction， and Qi and Yin deficiency syndrome. MethodA randomized， double-blind， positive-control， and multi-center clinical trial design method was used. 416 patients with diabetic retinopathy with blood stasis， collateral obstruction， and Qi and Yin deficiency syndrome in four test centers were included （the ratio of the treatment group to the control group was 3∶1）. On the basis of standardized hypoglycemic treatment， the treatment group was given both four Tongluo Mingmu capsules and two Calcium Dobesilate capsule agents three times a day， while the control group were given both two Calcium Dobesilate capsules and four Tongluo Mingmu capsule agents three times a day. The course of treatment was 12 weeks. The curative effect of Tongluo Mingmu capsules was evaluated by comparing the comprehensive curative effect of diabetic retinopathy， traditional Chinese medicine（TCM） syndrome score， corrected visual acuity， fundus changes， fundus fluorescence angiography， and other curative effect indexes before and after treatment in the two groups. At the same time， general examination， laboratory examination， and adverse events were performed to evaluate the safety of the drug. ResultThe baseline demographic data and disease characteristics of the treatment group and the control group were balanced and comparable， with the difference not statistically significant. After 12 weeks of treatment， the total effective rate of the comprehensive curative effect of diabetic retinopathy in the treatment group （61.0%， 189/310） was better than that in the control group （44.1%， 45/102）， and the difference was statistically significant （χ²=8.880， P<0.01）. The total effective rate of TCM syndromes in the treatment group （88.4%， 259/293） was better than that in the control group （69.9%， 65/93）， and the difference was statistically significant （χ²=17.927， P<0.01）. The disappearance rate of dry eyes （χ²=8.305）， dull complexion （χ²=4.053）， lassitude （χ²=10.267）， shortness of breath （χ²=8.494）， and dry stool （χ²=8.657） in the treatment group was higher than that in the control group， and the difference between the groups was statistically significant （P<0.05， P<0.01）. In terms of improving corrected visual acuity （χ²=8.382）， fundus changes （χ²=6.026） ， the treatment group was significantly better than the control group （P<0.05）. During the trial， the incidence of adverse events in the treatment group and the control group was 1.3% and 2.9%， respectively. There was no significant difference between the two groups. In addition， there were no serious adverse events and adverse events leading to withdrawal in both groups. ConclusionTongluo Mingmu capsules can improve the comprehensive curative effect of diabetic retinopathy and enhance the efficacy of TCM syndromes， visual acuity， fundus changes， and fundus fluorescein angiography， with great safety. Therefore， it can provide a new alternative therapeutic drug for patients with diabetic retinopathy.

The diagnosis of Parkinson's disease(PD)mainly relies on clinical appearance of patients,and the routine magnetic resonance scan has difficulty in finding the abnormality of that,which is not able to meet the requirement of early and precise quantification for the PD lesion area.Diffusion magnetic resonance imaging(dMRI)technique is a noninvasive method of examining the microstructure of living tissue.Diffusion weighted imaging(DWI)and diffusion tensor imaging(DTI)were the most widely used in clinical practice.However,the principles of them have limitation,which are not able to fully outline the microstructure of brain tissue.This paper discussed several new dMRI techniques that are able to make up conventional dMRI model,which included diffusion kurtosis imaging(DKI),neurite orientation dispersion and density imaging(NODDI)and diffusion tensor imaging index of perivascular space(DTI-ALPS).Moreover,this paper summarized the applications of above imaging technique in exploring the changes of the microstructure of brain tissue of PD patients,so as to provide more imaging information for the pathological changes of PD patients.

Objective:To compare the efficacy between dual stability constructs and modified Scott techniques for treatment of symptomatic spondylolysis in active adolescents.Methods:A retrospective study was conducted to analyze the clinical date of 64 active adolescents who had been treated for symptomatic spondylolysis at Department of Orthopedic Surgery, The 900th Hospital of Joint Logistic Support Force from January 2017 to October 2021. There were 59 males and 5 females with an age of (24.9±5.2) years. Responsible vertebral bodies were L 3 in 2 cases, L 4 in 10 cases, L 5 in 47 cases, and L 4 to L 5 in 5 cases; spondylolisthesis was accompanied in 9 cases. Depending on the surgical methods, the patients were divided into a dual stability constructs (pedicle screws and laminar screws) group (observation group, 31 cases) and a modified Scott group (control group, 33 cases). The 2 groups were compared in terms of operative time, bleeding volume, postoperative drainage volume, isthmus healing rate, rate of internal fixation failure, visual analogue scale (VAS) for low back pain, Japanese Orthopaedic Association (JOA) score and the good and excellent rate by JOA at postoperative 1 month, 3 months, and the last follow-up, and the incidence of complications. Results:There was no statistically significant difference in the preoperative general data between the 2 groups, indicating comparability ( P>0.05). All patients were followed up. The operation time in the observation group [(94.7±14.9) min] was significantly longer than that in the control group [(84.4±16.4) min] ( P=0.011), but there was no significant difference in intraoperative bleeding volume or postoperative drainage volume ( P>0.05). The healing rate of bilateral isthmi in the observation group was 93.5% (29/31), significantly higher than that in the control group [60.6% (20/33)], and the rate of internal fixation failure in the observation group (0) was significantly lower than that in the control group (12.1%, 4/33) ( P<0.05). At postoperative 1 month, 3 months, and the last follow-up, the VAS scores were significantly lower than the preoperative value in all patients while the JOA scores significantly higher ( P<0.05). At the last follow-up, in the observation group the VAS score [0 (0, 1.0)] was significantly lower than that in the control group [1(0, 2)], and the JOA score [(27.1±1.2) points] and the excellent and good rate by JOA [93.5% (29/31)] were significantly higher than those in the control group [(25.7±2.1) points and 75.8% (25/33)] ( P<0.05). In the control group, follow-ups revealed internal fixation failure in 4 cases due to the cable cutting out of the spinous processes, yielding a failure rate of 12.1%, while no internal fixation failure was observed in the observation group. Conclusions:Both dual stability constructs and modified Scott techniques can relieve the clinical symptoms of spondylolysis in active adolescents to various extents. However, dual stability constructs with pedicle screws and laminar screws may lead to a higher isthmus healing rate and better curative effects.

Objective:To investigate the characteristics and risk factors of local recurrence in resected pancreatic cancer.Methods:The clinicopathological data of 190 patients in whom recurrent sites can be identified after radical resection of pancreatic cancer from Sep 2013 to Aug 2020 at the Cangzhou Central Hospital were retrospectively analyzed. The survival time and clinicopathological characteristics of local recurrence were compared with those of other recurrence types. Cox risk regression model was used to analyze the risk factors of local recurrence.Results:The recurrence sites were local (49 cases, 25.8%), liver (53 cases, 27.9%), lung (35 cases, 18.4%), peritoneal (25 cases, 13.2%) and multiple sites (28 cases, 14.7%). Patients mRFS and mOS were 17.8 months and 30.9 months respectively. The clinicopathological features of patients with local recurrence were compared with those of other recurrence types [tumor diameter ( P=0.023), preoperative CA199 level ( P=0.021), peripancreatic nerve plexus invasion ( P=0.031), lymphovascular invasion ( P=0.004), surgical margin state ( P<0.001) and postoperative adjuvant chemotherapy ( P=0.038)]. Tumor diameter ( P=0.018), peripancreatic nerve plexus invasion ( P=0.002) and postoperative adjuvant chemotherapy ( P=0.004) were independent factors for local recurrence in resected pancreatic cancer, and only peripancreatic nerve plexus invasion was not associated with other recurrence types. Conclusions:Local recurrence in resected pancreatic cancer has important impact on the prognosis of patients. Peripancreatic nerve plexus invasion is an independent factor affecting local recurrence.

Objective:To evaluate the efficacy and safety of compound anisodine (CA) injection for anterior ischemic optic neuropathy (AION).Methods:Studies, which were randomized controlled trials (RCT) of CA with routine treatment for AION from PubMed, The Cochrane Library, Web of Science, CNKI, Wanfang database, Chinese science and technology journals full-text database and CD-ROM database of Chinese Biology Medical disc published from January 1978 to June 2021 were searched.Included studies were screened by two researchers independently based on inclusion and exclusion criteria.After data collection and quality assessment, a meta-analysis of included studies was performed with Revman 5.3 software.Results:Thirteen RCT were included, involving 1 636 eyes, with 829 eyes in treatment group and 807 eyes in control group without CA treatment.It was found that the total effective rate of treatment group was higher than that of control group ( OR=3.25, 95% CI: 2.47-4.28, P<0.01), and the best corrected visual acuity of AION patients after CA treatment was significantly better than that of control group when articles with high heterogeneity were excluded (MD=0.14, 95% CI: 0.11-0.17, P<0.01) or not (MD=0.14, 95% CI: 0.12-0.16, P<0.01).The mean defect of visual field was significantly smaller and the mean sensitivity of visual field was higher in treatment group than control group (MD=-2.58, 95% CI: -3.98--1.19, P<0.01; MD=3.49, 95% CI: 3.07-3.91, P<0.01). Conclusions:CA shows good efficacy in the treatment of AION.It can improve visual acuity and mean sensitivity of visual field, decrease mean defect of visual field.

Objective:To investigate the clinical application of Bushen-Liyan Decoction combined with low frequency pulse acupoint electrical stimulation in the treatment of swallowing dysfunction of stroke patients in convalescent stage. Methods:From January 2018 to March 2020, 108 stroke patients with swallowing dysfunction in the Shandong Provincial Third Hospital were selected. They were randomly divided into two groups with 54 in each according to the random number table method. The control group was given low frequency pulse acupoint electrical stimulation on the basis of conventional treatment, and the study group was given Bushen-Liyan Decoction combined with low frequency pulse acupoint electrical stimulation on the basis of conventional treatment. Both groups were treated for 4 weeks. The National Institutes of Health Neurological Deficit Score (NIHSS) was used to assess the degree of neurological impairment, the Wada Drinking Water Test was used to assess the degree of swallowing dysfunction, and Barthel Index was used to assess the ability of daily living. The serum levels of brain-derived neurotrophic factor (BDNF), insulin-like growth factor-1 (IGF-1), VEGF, IL-18, IL-23, SOD and MDA were detected by ELISA. The adverse reactions during the treatment were observed and the clinical efficacy was evaluated.Results:The total effective rate was 92.6% (50/54) in the study group and 72.2% (39/54) in the control group, with significant difference between the two groups( χ2=7.728, P=0.005). After treatment, NIHSS score and Wata’s drinking water test score of the study group were lower than those in the control group ( t=8.383, 9.740, P<0.001), Barthel index of the study group was significantly higher than that of the control group ( t=4.064, P<0.01). The serum BDNF [(8.42 ± 0.72) μg/L vs. (6.69 ± 0.63) μg/L, t=13.288], IGF-1[(142.20 ± 12.29) μg/L vs. (118.35 ± 10.73) μg/L, t=10.742] and VEGF[(151.29 ± 12.26) ng/L vs. (102.21 ± 9.77) ng/L, t=23.006] were significantly higher than those in control group( P<0.01); The serum levels of IL-18, IL-23 and MDA were significantly lower than those in the control group( t=7.892, 9.897, 6.206, P<0.001), and the level of SOD was significantly higher than that of the control group ( t=18.510, P<0.001). During the treatment, no serious adverse events occurred in both groups. Conclusion:Bushen-Liyan Decoction combined with low-frequency pulse acupoint electrical stimulation can reduce the clinical symptoms, inflammatory reaction and brain tissue damage of stroke patients with dysphagia, improve the swallowing function and improve the clinical efficacy.