Objective:To propose a deep learning（DL）-based ultrasound imaging auxiliary tool for automatic segmentation and recognition of the brachial plexus（BP），and to enhance the accuracy and safety of clinical procedures.Methods:It was a multicenter study that collected 773 healthy subjects from Peking University Third Hospital and its branch campuses，the Third Medical Center of the Chinese PLA General Hospital，and Shanghai Eighth People's Hospital between August 2024 and February 2025. Brachial plexus（BP）images in the interscalene groove were captured used high-frequency ultrasound by senior sonographers，a dataset comprising 1 289 standardized images were constructed and the improved model（CHA-TransUNet）was trained. The test set was input into 6 different models（CHA-TransUNet，R50-Unet，TransUnet，SegFormer，SwinUnet，MISSFormer）for segmentation. Segmentation accuracy was evaluated using metrics including the Dice similarity coefficient（DSC），95% Hausdorff distance（HD95）and mean intersection over union（mIoU），and was compared with the segmentation results of 3 ultrasound physicians with varying experience levels（junior physicians and senior physicians）to validate the model's segmentation efficacy.Results:The CHA-TransUNet model established based on a dataset of 1 289 standardized images achieved segmentation results for the BP with a DSC of 90.15%，mIoU of 91.02%，and HD95 of 8.08. Its accuracy was higher than other mainstream models（DSC：90.15% vs. 87.60%，87.77%，81.35%，84.78%，84.55%），significantly better than junior physicians（DSC：90.15% vs. 68.73%， Z=-127.76， P<0.001），and approached the level of senior physician（DSC：90.15% vs. 86.15%， Z=-31.33， P=0.549）. The model demonstrated superior boundary recognition in complex anatomical structures（e.g.，C6/C7 nerve roots）compared to ultrasound physicians（junior and senior）（HD95：8.08 vs. 26.34，17.44，56.80）. Conclusions:This study proposes an analysis model for BP ultrasound images，CHA-TransUNet. This model achieves segmentation and recognition of the BP with relatively complex pathways and structures. The model exhibits high accuracy and stability，outperforming current mainstream network models and junior physicians while approaching the performance level of senior physicians. It assists junior physicians or trainees in more accurately identifying and localizing the BP.

Objective:To explore the binding characteristics of N, N-diethyl-2-(2-(4-(2- 18F-fluoroethoxy)phenyl)-5, 7-dimethylpyrazolo[1, 5-a]pyrimidin-3-yl)acetamide ( 18F-DPA-714) and ( R)- N-sec-butyl- N-methyl-4-(3-( 18F-trifluoromethyl)phenyl)quinazoline-2-carboxamide ( 18F-TFQC) to the single nucleotide polymorphisms of the 18×10 3 translocator protein (TSPO), and to evaluate the imaging efficacy and feasibility of those 2 molecular probes in neuroinflammation rat models. Methods:To test the selectivity of 18F-DPA-714 and 18F-TFQC for TSPO polymorphisms, the wild-type (high-affinity binding, HAB) and mutant (low-affinity binding, LAB) sequences of the human TSPO gene were transfected into 293T cells respectively. A competitive inhibition assay was carried out with N-methyl- N-(1-methylpropyl)-1-(2-chlorophenyl)-3-isoquinoline carboxamide (PK11195) as an inhibitor to determine the binding affinities of 2 probes to TSPO polymorphisms. Rat neuroinflammation models ( n=6) were established using lipopolysaccharide. Three days after modeling, small animal PET/CT imaging was performed using 18F-DPA-714 and 18F-TFQC, respectively, to observe and compare the uptake of the tracers, and the ratio of SUV mean of the right striatum to SUV mean of the left striatum (SUVR) was calculated. After the imaging, the expression and distribution of microglia and TSPO were detected by tissue immunofluorescence. Repeated-measures analysis of variance was used to analyze the SUVR data of different groups. Results:The inhibition constants ( Ki) of 18F-TFQC on 293T-LAB and 293T-HAB cells were 23.51 and 14.60 nmol/L, respectively, with a Ki LAB/ Ki HAB ratio of 1.61, indicating low sensitivity to TSPO single nucleotide polymorphisms. The Ki of 18F-DPA-714 for binding to 293T-LAB and 293T-HAB cells were 45.23 and 6.47 nmol/L, respectively, with a Ki LAB/ Ki HAB ratio of 6.99. Small animal PET/CT imaging demonstrated that specifically uptake of both probes could be found in neuroinflammatory lesions. The overall SUVR of 18F-DPA-714 in the lesions within 60minutes was slightly higher than that of 18F-TFQC, but no significant difference was observed ( F values: inter-group 0.40, time effect 0.30, cross-effect 0.03; all P>0.05). Conclusions:Compared with 18F-DPA-714, 18F-TFQC is less sensitive to TSPO gene polymorphisms, thus being more suitable for clinical application and promotion. It holds promise for the early identification of neuroinflammation and the efficacy monitoring of anti-inflammatory drug treatments.

The translocator protein (TSPO) positron emission tomography (PET) can noninvasively detect neuroinflammation associated with epileptogenesis and epilepsy. This study explored the role of the TSPO-targeting radioligand [¹⁸F]F-TFQC, an m-trifluoromethyl ER176 analog, in the PET neuroimaging of epileptic rats. Initially, [¹⁸F]F-TFQC was synthesized with a radiochemical yield of 8%-10% (EOS), a radiochemical purity of over 99%, and a specific activity of 38.21 ± 1.73 MBq/nmol (EOS). After determining that [¹⁸F]F-TFQC exhibited good biochemical properties, [¹⁸F]F-TFQC PET neuroimaging was performed in epileptic rats at multiple time points in various stages of disease progression. PET imaging showed specific [¹⁸F]F-TFQC uptake in the right hippocampus (KA-injected site, i.e., epileptogenic zone), which was most pronounced at 1 week (T/NT 1.63 ± 0.21) and 1 month (T/NT 1.66 ± 0.20). The PET results were further validated using autoradiography and pathological analysis. Thus, [¹⁸F]F-TFQC can reflect the TSPO levels and localize the epileptogenic zone, thereby offering the potential for monitoring neuroinflammation and guiding anti-inflammatory treatment in patients with epilepsy.

Objective:To explore the binding characteristics of N, N-diethyl-2-(2-(4-(2- 18F-fluoroethoxy)phenyl)-5, 7-dimethylpyrazolo[1, 5-a]pyrimidin-3-yl)acetamide ( 18F-DPA-714) and ( R)- N-sec-butyl- N-methyl-4-(3-( 18F-trifluoromethyl)phenyl)quinazoline-2-carboxamide ( 18F-TFQC) to the single nucleotide polymorphisms of the 18×10 3 translocator protein (TSPO), and to evaluate the imaging efficacy and feasibility of those 2 molecular probes in neuroinflammation rat models. Methods:To test the selectivity of 18F-DPA-714 and 18F-TFQC for TSPO polymorphisms, the wild-type (high-affinity binding, HAB) and mutant (low-affinity binding, LAB) sequences of the human TSPO gene were transfected into 293T cells respectively. A competitive inhibition assay was carried out with N-methyl- N-(1-methylpropyl)-1-(2-chlorophenyl)-3-isoquinoline carboxamide (PK11195) as an inhibitor to determine the binding affinities of 2 probes to TSPO polymorphisms. Rat neuroinflammation models ( n=6) were established using lipopolysaccharide. Three days after modeling, small animal PET/CT imaging was performed using 18F-DPA-714 and 18F-TFQC, respectively, to observe and compare the uptake of the tracers, and the ratio of SUV mean of the right striatum to SUV mean of the left striatum (SUVR) was calculated. After the imaging, the expression and distribution of microglia and TSPO were detected by tissue immunofluorescence. Repeated-measures analysis of variance was used to analyze the SUVR data of different groups. Results:The inhibition constants ( Ki) of 18F-TFQC on 293T-LAB and 293T-HAB cells were 23.51 and 14.60 nmol/L, respectively, with a Ki LAB/ Ki HAB ratio of 1.61, indicating low sensitivity to TSPO single nucleotide polymorphisms. The Ki of 18F-DPA-714 for binding to 293T-LAB and 293T-HAB cells were 45.23 and 6.47 nmol/L, respectively, with a Ki LAB/ Ki HAB ratio of 6.99. Small animal PET/CT imaging demonstrated that specifically uptake of both probes could be found in neuroinflammatory lesions. The overall SUVR of 18F-DPA-714 in the lesions within 60minutes was slightly higher than that of 18F-TFQC, but no significant difference was observed ( F values: inter-group 0.40, time effect 0.30, cross-effect 0.03; all P>0.05). Conclusions:Compared with 18F-DPA-714, 18F-TFQC is less sensitive to TSPO gene polymorphisms, thus being more suitable for clinical application and promotion. It holds promise for the early identification of neuroinflammation and the efficacy monitoring of anti-inflammatory drug treatments.

Objective:To propose a deep learning（DL）-based ultrasound imaging auxiliary tool for automatic segmentation and recognition of the brachial plexus（BP），and to enhance the accuracy and safety of clinical procedures.Methods:It was a multicenter study that collected 773 healthy subjects from Peking University Third Hospital and its branch campuses，the Third Medical Center of the Chinese PLA General Hospital，and Shanghai Eighth People's Hospital between August 2024 and February 2025. Brachial plexus（BP）images in the interscalene groove were captured used high-frequency ultrasound by senior sonographers，a dataset comprising 1 289 standardized images were constructed and the improved model（CHA-TransUNet）was trained. The test set was input into 6 different models（CHA-TransUNet，R50-Unet，TransUnet，SegFormer，SwinUnet，MISSFormer）for segmentation. Segmentation accuracy was evaluated using metrics including the Dice similarity coefficient（DSC），95% Hausdorff distance（HD95）and mean intersection over union（mIoU），and was compared with the segmentation results of 3 ultrasound physicians with varying experience levels（junior physicians and senior physicians）to validate the model's segmentation efficacy.Results:The CHA-TransUNet model established based on a dataset of 1 289 standardized images achieved segmentation results for the BP with a DSC of 90.15%，mIoU of 91.02%，and HD95 of 8.08. Its accuracy was higher than other mainstream models（DSC：90.15% vs. 87.60%，87.77%，81.35%，84.78%，84.55%），significantly better than junior physicians（DSC：90.15% vs. 68.73%， Z=-127.76， P<0.001），and approached the level of senior physician（DSC：90.15% vs. 86.15%， Z=-31.33， P=0.549）. The model demonstrated superior boundary recognition in complex anatomical structures（e.g.，C6/C7 nerve roots）compared to ultrasound physicians（junior and senior）（HD95：8.08 vs. 26.34，17.44，56.80）. Conclusions:This study proposes an analysis model for BP ultrasound images，CHA-TransUNet. This model achieves segmentation and recognition of the BP with relatively complex pathways and structures. The model exhibits high accuracy and stability，outperforming current mainstream network models and junior physicians while approaching the performance level of senior physicians. It assists junior physicians or trainees in more accurately identifying and localizing the BP.

Objective:To explore whether the specific synaptic vesicle glycoprotein 2A (SV2A) targeted imaging agent ( R)-4-(3-fluoro-5-(fluoro- 18F)phenyl)-1-((3-methylpyridin-4-yl)methyl)pyrrolidin-2-one ( 18F-SDM-8) can be used to detect epileptic foci. Methods:Twenty male Sprague-Dawley (SD) rats (8-9 weeks) were injected with 1.2 μl of kainic acid (16 rats in the epilepsy group) or saline (4 rats in the control group) into the right hippocampus. 18F-SDM-8 and 18F-FDG mircoPET/CT imaging were respectively performed at 1-2 d (acute phase), 6-7 d (incubation period) and 45-60 d (chronic phase) after the seizure. Asymmetric index (AI) was used to evaluate the epileptic foci identify ability of 18F-SDM-8. Paired t test, Mann-Whitney U test and Pearson correlation analysis were used to analyze data. Results:In the three periods of 18F-SDM-8 imaging, the differences of AI of hippocampus between the epilepsy group and control group were statistically significant ( z values: from -2.64 to 2.67, all P<0.05). Both imaging agents had asymmetric uptake in the epilepsy group (right was lower than left), and the decrease in the medial right temporal lobe was the most significant. The pathological staining results were consistent with the imaging results. In the chronic phase of the epilepsy group, the differences of 18F-SDM-8 SUV mean (right versus left) in each brain area of interest were statistically significant ( t value: from -33.40 to -5.60, all P<0.05). The asymmetric uptake of the two imaging agents in the hippocampus had a better correlation ( r=0.97, P=0.001), and the AI of 18F-SDM-8 ((34.2±8.4)%) in this area was 1.4 times higher than that of 18F-FDG ((24.6±4.7)%). Conclusions:18F-SDM-8 PET is a promising method to test the level of SV2A. It can reflect the changes of SV2A in the rat epilepsy model induced by intrahippocampal injection of kainic acid, and improve the sensitivity of molecular imaging for epileptic foci.

Objective:To explore the effect of mobile health (mHealth) management system on dietary behavior of patients with coronary heart disease (CHD) .Methods:Using the convenient sampling method, CHD patients who were hospitalized in Department of Cardiology of Beijing Anzhen Hospital and Beijing Chaoyang Hospital of Capital Medical University were selected as the research objects from June 2019 to January 2020. The research objects were randomly divided into the intervention group and the control group using SAS software. Finally, 68 cases in the intervention group and 66 cases in the control group completed the study. The intervention group received mhealth management, while the control group received routine management, and they were followed up to 1 month after discharge. The proportion of patients with the low-salt and low-oil diet, the intake of salt, oil, staple food, vegetables, fruits, fish, shrimp, poultry and eggs, BMI and blood pressure were compared between the two groups.Results:At 1 month follow-up, the intervention group was better than the control group in terms of low salt and low oil diet, salt intake, oil intake and vegetable intake compliance, and the differences were statistically significant ( P<0.05) . At 1 month follow-up, there were no statistically significant differences in BMI and blood pressure between two groups ( P>0.05) . Conclusions:The mobile health system can effectively increase the proportion of patients with a low-salt and low-oil diet, and improve the intake compliance of salt, edible oil and vegetables in patients with CHD. In future studies, the intervention time should be extended to further explore the long-term intervention effect of mobile health and give full play to the role of mobile health in continuous behavior management of patients with CHD after discharge.

Objective To summarize the key points of nursing care in reservation of catheters in 11 cases of nasopharyngeal carcinoma patients who received concurrent chemo-radiotherapy with symptomatic thrombosis after ultrasound-guided implantation of double-lumen PICCs(PowerPICC).Methods From January,2014 to December,2015,totally 11 cases with symptomatic venous thromboembolism were identified among 109 cases of nasopharyngeal carcinoma patients receiving concurrent radiotherapy and chemotherapy,and observation and nursing care were provided at the early stage of thrombosis and during thrombosis.Results All double-lumen PICCs were reserved,and no recurrence or aggravation of thrombosis was recorded to the end of the treatment.The duration of carrying PICCs was 67～89(77.45±6.65) days.Conclusion With careful treatment and nursing,PICC catheter-related complications can be reduced and the duration of carrying catheters can be prolonged,which leads to accomplishment of the treatment plan for patients.

ObjectiveTo investigate methods of early postoperative rehabilitation and its effects on patients with bicondylar tibial plateau fracture treated by modified dual plating.Methods66 patients with bicondylar tibial plateau fracture from Tianjin Hospital were divided into 2 groups according to the starting time for postoperative rehabilitation: early rehabilitation group (n=30), convalescents rehabilitation group (n=36). The two groups had the same treatment since the fourth week postoperative. Tibial plateau angle (TPA) and posterior slope angle (PA) of tibial plateau were measured to evaluate the stability of proximal tibial and knee alignment. The Hospital for Special Surgery (HSS) score system and knee joint range of motion (ROM) were used as the parameters for knee function.ResultsThe mean value of TPA and PA had no significant changes, compared to the final follow-up and postoperative X-ray films(P>0.05); the differences between the two groups were statistically significant in the mean HSS score and mean value of knee flexion and extention ROM(P<0.01 or 0.05). There was significantly positive correlation between starting time and the total number of treatment was significant(P<0.01), while negative correlation with flexion ROM of knee joint and the HSS score(P<0.01).ConclusionThe correlation between rehabilitation intervention timing and functional recovery for patients with bicondylar tibial plateau fracture treated by modified dual plating is significant, early individual and systematic rehabilitation therapy can effectively improve the prognosis features of patients.

Objective To construct a short hairpin RNA (shRNA) adenovirus vector targeting Aktl (protein kinase B1, PKBI/Aktl) and cyclooxygenase-2 (COX-2) and study its effects on the invasion and metastasis of SGC-7901 human gastric adenocarcinoma cells. Methods Aktl and COX-2 shRNA expression frames were subcloned to pGSadeno adenovirus vector by homologous recombination technology to construct pGSadeno-Aktl+COX-2 (rAdS-A+C) vector. After screening and amplification, the recombinant adenovirus vector was digested with PacI and transfected into SGC-7901 cells, the titer and transfection efficiency were detected by fluorescent microscopy. Aktl and COX-2 mRNA and protein expression was identified by real-time PCR and Western blot. MMP-2 and MMP-9 contents in control group SGC-7901、rAd5-HK and treatment group rAdS-A+C were detected by ELISA assay and transwell assay analyzed cell invasion and metastasis ability. Results Adenovirus vector rAdS-A+C was successfully constructed and it dramatically down-regulated Aktl and COX-2 mRNA and protein expression in SGC-7901 gastric cancer cells. MMP-2 and MMP-9 contents in treatment group rAd5-A+C were respectively (39.7± 1.7) ng/ml, (31.3±3.6) ng/ml, and they were lower than those in control group SGC-7901 (278.4± 15.5) ng/ml, (225.4±15.1) ng/ml and rAd5-HK (275.5±2.1) ng/ml, (226.0±23.3) ng/ml (P= 0.01, P=0.021). Transwell assay showed treatment group rAd5-A+C significantly inhibited the invasion and metastasis of SGC-7901 gastric adenocacinoma cells. Conclusions Adenovirus-mediated targeting Aktl and COX-2 shRNA can inhibit the invasion and metastasis of SGC-7901 human gastric adenocarcinoma cells.