BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

Studies have suggested that the nucleus accumbens (NAc) is implicated in the pathophysiology of major depression; however, the regulatory strategy that targets the NAc to achieve an exclusive and outstanding anti-depression benefit has not been elucidated. Here, we identified a specific reduction of cyclic adenosine monophosphate (cAMP) in the subset of dopamine D1 receptor medium spiny neurons (D1-MSNs) in the NAc that promoted stress susceptibility, while the stimulation of cAMP production in NAc D1-MSNs efficiently rescued depression-like behaviors. Ketamine treatment enhanced cAMP both in D1-MSNs and dopamine D2 receptor medium spiny neurons (D2-MSNs) of depressed mice, however, the rapid antidepressant effect of ketamine solely depended on elevating cAMP in NAc D1-MSNs. We discovered that a higher dose of crocin markedly increased cAMP in the NAc and consistently relieved depression 24 h after oral administration, but not a lower dose. The fast onset property of crocin was verified through multicenter studies. Moreover, crocin specifically targeted at D1-MSN cAMP signaling in the NAc to relieve depression and had no effect on D2-MSN. These findings characterize a new strategy to achieve an exclusive and outstanding anti-depression benefit by elevating cAMP in D1-MSNs in the NAc, and provide a potential rapid antidepressant drug candidate, crocin.

Objective:To analyze the general chapter for lotions 0127 of the Chinese Pharmacopoeia 2020 Vol-ume Ⅳ,and discuss how to improve the general technical requirements of lotions 0127 in the Chinese Pharma-copoeia.Methods:By comparing the general chapter for lotions in domestic and international pharmacopoe-ias,the definition,classification,process,storage and corresponding inspection requirements were analyzed.Results and Conclusions:The general chapter for lotions 0127 of the Chinese Pharmacopoeia should be revised,including improvement of the definition,increasement of forms of preparations,and expansion of included varieties,so as to promote scientific regulation for drugs and exhibit a guiding role of the Chinese pharmacopoeia in drug control.

Objective To summarize the experience in anesthesia management for cesarean section(CS) after failed labor in women with combined spinal and epidural analgesia and to provide evidence for more effective and safer clinical practice.Methods We included 58 singleton pregnant women who received spinal and epidural analgesia during labor but converted to CS at Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology from October 2017 to October 2018.When CS was indicated,10 ml of 0.125％ ropivacaine was given through the epidural catheter immediately,and 5 ml of 1％ lidocaine was given when arriving at the operating room.Five minutes later,10 ml of 1％ ropivacaine was administered.General anesthesia would be required when epidural anesthesia was considered to be failed 10 min after the usage of ropivacaine.Clinical datas were retrospectively reviewed.The management and effects of anesthesia for CS,maternal and neonatal outcomes were described.Results Of the 58 patients when CS was indicated during labor,5.2％(3/58) received general anesthesia immediately.Among the other 94.8％ (55/58) who received epidural anesthesia,3.6％(2/55) converted to general anesthesia later.Adverse effects of epidural anesthesia included reduced mean artery pressure (8/53,15.1％),nausea and vomiting (3/53,5.7％).For the neonatal Apgar score at 1 min,one out of the 58 babies (1.7％) was between 0 and 3,ten (17.2％) between 4 and 7,and 47 (81.0％) between 8 and 10.The two babies (3.4％) with Apgar score between 4 and 7 at 5 min were both referred to the department of Neonatology,and the rest 56 neonates (96.6％) scored 8-10.Conclusions Spinal and epidural analgesia in labor can be safely and effectively adjusted to anesthesia for CS.

Objective To investigate the clinical imaging features and immune manifestations of clear cell carcinoma and papillary renal cell carcinoma at the same time in one side of the kidney,in order to increase the awareness of the disease and improve the clinical diagnosis and treatment.Methods This study retrospectively analyzed the clinical data of 7 patients with unilateral renal clear cell carcinoma combined with papillary cell carcinoma admitted to the Department of Urology,People's Hospital of Xinjiang Uygur Autonomous Region from January 2007 to June 2017.The general data,images performance,surgical methods,immunohistochemical results of the patients were observed.Results In 7 patients with multiple tumors of ipsilateral kidney,6 were males,accounting for 85.71％ (6/7);1 female,accounting for 14.29％ (1/7).Ages 45-78 years old,average 63 years old;4 cases were found by physical examination,accounting for 57.14％ (4/7);2 cases were admitted to hospital with obvious oncological symptoms,accounting for 28.57％ (2/7);1 case of renal nonfunction,accounting for 14.29％ (1/7).The tumor was located in 6 cases of left kidney,accounting for 85.71％ (6/7),and the tumor was located in 1 case of right kidney,accounting for 14.29％ (1/7).Preoperative imaging examination clearly indicated multiple tumors in 3 cases,and the remaining 4 cases only showed placeholders.Lesions,of which the maximum diameter of the tumor 7.5 cm,2 patients with a diameter of ＞ 4 cm;4 cases of radical renal resection and partial nephrectomy;3 cases of clear cell carcinoma Fuhrman Ⅰ grade,Fuhrman Ⅱ grade 5 cases postoperative pathology report There were 4 cases of papillary renal cell carcinoma in Fuhrman grade Ⅰ and 3 cases in Fuhrman grade Ⅱ,all of them were type Ⅰ cell carcinoma.Conclusions Ipsilateral kidney associated with papillary renal cell carcinoma and clear cell carcinoma is rare clinically,its formation is unknown.Readings should be repeated before surgery.If you can identify multiple tumors,you will be able to choose the surgical procedure.Partial nephrectomy or radical nephrectomy may be performed depending on the size of the tumor.Radical nephrectomy is recommended.

Objective To evaluate the clinical effect of iliac artery interventional chemotherapy combination with bladder perfusion for high level of bladder urothelial carcinoma after transurethral bladder tumor electricity cut operation for those who intend to retain the bladder.Methods From February 2010 to December 2016,a total of 74 cases high-grade urothelial carcinoma of the bladder,giving chemotherapy and suitted support were retrospectively analyzed.The patients were divided into two groups according to different chemotherapy methods.Artery perfusion chemotherapy group:43 cases treated by transurethral bladder tumor cutting + bilateral iliac artery interventional perfusion chemotherapy + bladder perfusion in combination therapy;Intravenous chemotherapy group:31 cases underwent transurethral bladder tumor cut method + cisplatin,vein gemcitabine + bladder perfusion chemotherapy combined therapy.According to the ((x) ± s) between the groups.The method of count data is represented by chi-square test.Kaplan-meier method was used to describe the survival time of two groups of patients after operation.Results Two groups of patients with postoperative there was no statistically significant difference in disease-free survival time.In the bone marrow suppression (x2 =4.956,P =0.029);gastrointestinal tract reaction (x2 =5.912,P =0.012);dermatitis,mucositis (x2 =4.276,P =0.013),etc.,has the advantages in the iliac artery interventional perfusion chemotherapy (P ＜ 0.05).Conclusion Via efficient iliac arterial infusion chemotherapy in patients with venous chemotherapy patients,to the retention of the bladder has a certain value,and the incidence of adverse reactions after surgery is low,have certain advantages.

BACKGROUNDThe aim of this research was to analyze the perioperative factors of regular hepatectomy and irregular hepatectomy. The superiority of the clinical application of the two methods was compared in the perioperative period.

METHODSFrom 1986 to 2011, 1798 patients underwent consecutive liver resections with regular hepatectomy and irregular hepatectomy at the Air Force General Hospital of People's Liberation Army and the General Hospital of Chinese People's Liberation Army. Their medical documentation was investigated retrospectively.

RESULTSIn patients on whom regular hepatectomy and irregular hepatectomy were performed, there was no significant difference in perioperative blood loss, complications, in-hospital mortality, hospital stay, and so on. But in regular hepatectomy, operating time was an independent risk factor (P < 0.001, OR = 1.004).

CONCLUSIONSThere was no significant difference between the perioperative risk of regular hepatectomy and that of irregular hepatectomy.

Objective To introduce a new spinal surgery navigation system based on structured light scanning(structured light navigation system,SLNS),and to compare accuracy of pedicle screw placement using SLNS,CT based navigation system and free hand technique.Methods Thirty-two calf vertebral pedicles,40 and 32 were placed with pedicle screws using SLNS,CT navigation system,and free hand technique,respectively.The pedicle breakage ratio,transverse section angle (TSA),sagittal section angle (SSA),screw offset deviation of each screw were detected according to CT images.Results The pedicle breakage ratio of pedicle screws in SLNS group,CT navigation group and free hand group were 6%(2./32),5%(2/40),25%(8/32),respectively.The difference between each of the two navigation groups and free hand group were statistically significant regarding frequency of screw misplacement.The difference between the two navigation systems was not obvious.The average SSA error and TSA error in SLNS group were 2.58°±2.74° and 4.26°±5.20°.For CT navigation group,they were 2.95°±2.61° and 3.13°±-2.75°.There was no statistical difference between the two navigation methods for the SSA or TSA error.There was no statistical difference in offset deviation among SLNS group,CT navigation group and free hand group.Conclusion The SLNS is a new and practical navigation system,which has similar accuracy with CT navigation system.

Objective To construct recombinant full length genotype B and C hepatitis B virus (HBV)and to examine HBV DNA replication and hepatitis B surface antigen(HBsAg),hepatitis B e antigen(HBeAg)expressions in Huh7 cells. Methods The full length genotype B and C HBV DNA were extracted and amplified from two HBV infected patients. The recombinant plasmids were constructed by inserting the amplified HBV fragments into the eukaryotic expression vector,pHY106,which were then transfected into Huh7 cells. The cells transfected with blank pHY106 vector were used as control. HBV DNA replication at 72 hours of transfection was detected by Southern blot. The HBV DNA levels in Huh7 cells at 24,48,72,96 and 120 hours of transfection were determined by real-time polymerase chain reaction(PCR).Meanwhile, the HBsAg and HBeAg expression levels in the supernatants at 24,48,72,96 and 120 hours were determined by enzyme linked immunosorbent assay(ELISA).Results The recombinant plasmids expressing genotype B or C HBV DNA were successfully constructed.The HBV replicative intermediates in HBV core particles,including rcDNA dsDNA and ssDNA,were detected by Southern blot.HBV DNA level could reach 8 lg copy/mL which was by real-time PCR. HBsAg and HBeAg levels determined by ELISA peaked at 72 hours after transfection and then declined gradually. Conclusions The recombinant plasmids inserted with genotype B or C HBV DNA are constructed successfully, which can express high levels of HBsAg and HBeAg in Huh7 cells. This system provides a platform for studying the pathogenesis of B and C genotype HBV, the interaction between HBV and host, as well as exploiting new drugs against HBV.