BACKGROUND:Socket-shield technique can effectively maintain labial soft and hard tissues,but the incidence of postoperative complications such as exposure and displacement of root shield is relatively high.It is speculated that the root shield may be exposed and displaced due to excessive load after long-term function of dental implants. OBJECTIVE:Through three-dimensional finite element analysis,we aim to study the influence of varying root shield thicknesses on the stress distribution,equivalent stress peaks,and displacement in the root shield,periodontal ligaments,implant,and surrounding alveolar bone under normal occlusal loading.We also attempt to analyze the correlation between the thickness of the root shield and occurrence of mechanical events such as root shield exposure,displacement,and fracture. METHODS:Cone-beam CT data of a patient who met the indication standard of socket-shield technique for maxillary central incisor were retrieved from database.Reverse engineering techniques were used to build models of the maxillary bone and root shield,while forward engineering was used to create models for the implant components based on their parameters.Models depicting various root shield thicknesses(0.5,1.0,1.5,and 2.0 mm)were created using Solidworks 2022 software.ANSYS Workbench 2021 software was then used to simulate and analyze the effects of varying root shield thicknesses on stress distribution,equivalent stress peaks,and displacement of the root shields,periodontal ligaments,implants,and surrounding alveolar bone under normal occlusion. RESULTS AND CONCLUSION:(1)In all root shield models,the stress was concentrated on the palatal cervical side,both sides of the edges and the lower edge of the labial side.As the thickness of the root shield increased,the equivalent stress peak and displacement showed a decreasing trend.The 0.5 mm thickness model produced a stress concentration of 176.20 MPa,which exceeded the yield strength(150 MPa)of tooth tissue.(2)The periodontal ligament stress in each group was concentrated in the neck margin and upper region.With the increase of root shield thickness,the equivalent stress peak and displacement of periodontal ligament showed a decreasing trend.(3)Implant stress in all models was concentrated in the neck of the implant and the joint of the implant-repair abutment,and the labial side was more concentrated than the palatal side.With the increase of root shield thickness,the equivalent stress peak of the implant in the model showed an increasing trend.(4)In each group of models,stress of cortical bone concentrated around the neck of the implant and the periphery of the root shield,and the labial side was more concentrated than the palatal side.With the increase of the thickness of the root shield,the equivalent stress peak around the root shield decreased;the peak value of the equivalent stress of the bone around the neck of the implant showed an increasing trend.In the model,the stress of cancellous bone was mainly concentrated around the neck of the lip of the implant,the top of the thread,the root tip and the lower margin of the root shield,and the labial side was more concentrated than the palatal side.With the increase of the thickness of the root shield,the peak value of the equivalent stress of the bone around the root shield in the model showed a decreasing trend.The minimum principal stress of cortical bone in each group of models was concentrated around the neck of the implant,exhibiting a fan-shaped distribution.As the thickness of the root shield increased,the minimum principal stress of cortical bone showed an increasing trend.(5)These results indicate that different thicknesses of the root shield have different biomechanical effects.The root shield with a thickness of 0.5 mm is easy to fracture.For patients with sufficient bone width,the root shield with a thickness of 2.0 mm is an option to reduce the risk of complications such as root shield exposure,fracture,and displacement.Meanwhile,it should be taken into account to protect the periodontal ligament in the preparation process,and rounding treatments ought to be carried out on both sides and the lower edge of the root shield.

Objective To improve the national standardized method for determining biphenyl in workplace air, which was based on activated carbon tube sampling, carbon disulfide desorption, and gas chromatography, by developing a method using GDX-502 tubes for sampling, toluene for desorption, and gas chromatography. Methods Workplace air samples were collected using GDX-502 sampling tubes and desorbed with toluene, followed by determination with gas chromatography. Results The improved method demonstrated good linearity for biphenyl concentrations ranging from 0.33 to 330.00 mg/L, with a correlation coefficient of 0.999 9. The detection limit and lower limit of quantification were 0.06 and 0.21 mg/L, and the minimum detection concentration and minimum quantification concentration were 0.04 and 0.14 mg/m³ (based on 1.5 L air sample volume), respectively. The average desorption efficiency ranged from 96.6% to 101.1%. The within-run and between-run relative standard deviations were 0.6%-1.4% and 1.4%-3.3%, respectively, with 100.0% sampling efficiency. Samples remained stable for at least 14 days at room temperature. Conclusion The improved method for biphenyl detection demonstrates rapid and accurate performance, with the advantages of low detection limits and high sampling and desorption efficiency.

Objective To establish a liquid chromatography method for the simultaneous determination of 13 aromatic amine compounds (AAs) in workplace air. Methods A total of 13 AAs in both vapor and aerosol phases were collected in workplace air using a new GDH-6 sampling tube. Samples were desorbed and eluted with methanol, separated using a Symmetry Shield™ RP₁₈ reversed-phase liquid chromatography column, and detected with a diode array detector. Quantification was performed using an external standard method. Results The linear range of the 13 AAs measured by this method was 0.02-373.60 μg/L with the correlation coefficients greater than 0.999 0. The minimum detection concentration was 0.09-14.37 μg/m³, and the minimum quantitative concentration was 0.31-47.90 μg/m³ (both calculated based on sampling 15.0 L of air and 3.0 mL of elution volume). The average desorption and elution efficiency ranged from 97.46% to 101.23%. The within-run relative standard deviation (RSD) was 0.10%-5.99%, and the between-run RSD was 0.17%-2.71%. Samples could be stably stored in sealed conditions at 2-8 ℃ for more than seven days. Conclusion This method is suitable for the simultaneous determination of 13 AAs in workplace air, including both vapor and aerosol phases.

Objective To establish a method for simultaneous determination of four high-molecular-weight thiol derivatives (TDs) in workplace air by gas chromatography. Methods The four kinds of vapor-phase macromolecular TDs (1-pentanethiol, 1-hexanethiol, 1-benzyl mercaptan, and n-octanethiol) in the workplace air were collected using the GDH-1 air sampling tubes, desorbed with anhydrous ethanol, separated on a DB-FFAP capillary column, and determined by flame ionization detector. Results The quantitation range of the four TDs was 0.30-207.37 mg/L, with the correlation coefficients greater than 0.999 00. The minimum detection mass concentrations and minimum quantitation mass concentrations were 0.18-0.32 and 0.60-1.05 mg/m³, respectively (both calculated based on the 1.5 L sample and 3.0 mL desorption solvent). The mean desorption efficiencies ranged from 87.07% to 103.59%. The within-run and between-run relative standard deviations were 1.92%-8.22% and 1.89%-8.45%, respectively. The samples can be stored at room temperature or 4 ℃ for three days and up to 7 days at -18 ℃. Conclusion This method is suitable for the simultaneous determination of four vapor-phase TDs in workplace air.

Objective:To construct the over-expressed lentivirus vector of PRDM5 gene and establish the Neuro-2a cells stably transfected PRDM5,and to provide the basis evidence for exploring the effect of PRDM5 in pathogenesis of ischemic stroke(IS).Methods:The sequence of PRDM5 was searched and designed based on NCBI.The PRDM5 gene was amplified by PCR and ligated with the lentiviral vector GV492 digested by BamH Ⅰ and Age Ⅰ restriction enzymes to form the GV492-PRDM5 over-expression recombinant plasmid.The positive clones with similar length and size to the target gene fragment were screened by PCR and sent to Shenggong Bioengineering(Shanghai)Co.Ltd.for identification.The correctly-sequenced GV492-control plasmid and GV492-PRDM5 over-expression recombinant plasmid were transfected into the HEK293T cells,respectively.After 48 h of transfection,the lentiviruses were collected by centrifugation,and they were GV492-control lentivirus and GV492-PRDMS over-expression lentivirus;the titers of these two lentiviruses were determined by lentiviral titer assay.The Neuro-2a cells were divided into GV492-control group and GV492-PRDM5 group,and then infected with GV492-control lentivirus and GV492-PRDM5 over-expression lentivirus,respectively,with a lentivirus multiplicity of infection(MOI)of 100.The Neuro-2a cells successfully infected with GV492-control lentivirus and GV492-PRDM5 over-expression lentivirus were screened with puromycin(10 mng-L-1)after 72 h of infection.The growth status and the expression of green fluorescence protein of Neuro-2a cells in GV492-control group and GV492-PRDM5 group were observed by fluorescence microscope.The expression levels of PRDM5 mRNA and PRDM5 protein in the Neuro-2a cells in two groups were detected by real-time fluorescence quantitative RCR(RT-qPCR)and Western blotting methods.Results:The PCR results showed that the length of the positive transformant of GV492-PRDM5 recombinant plasmid was about 684 bp,and the gene sequence of GV492-PRDM5 over-expression recombinant plasmid was consistent with the designed and synthesized PRDM5 over-expression sequence.The titers of GV492-control lentivirus and GV492-PRDM5 over-expression lentivirus were both 2.5×108TU·mL-1 The Neuro-2a cells in GV492-control group and GV492-PRDM5 group grew well,and the expressions of green fluorescence protein were found under fluorescence microscope.The RT-qPCR results showed that the expression level of PRDM5 mRNA in the Neuro-2a cells in GV492-PRDM5 group was significantly increased compared with GV492-control group(P＜0.01).The Western blotting results showed that the specific bands appeared in the Neuro-2a cells in GV492-control group and GV492-PRDM5 group with a relative molecular weight of 75 000;compared with GV492-control group,the expression level of PRDM5 protein in the Neuro-2a cells in GV492-PRDM5 group was increased(P＜0.01).Conclusion:The over-expression lentivirus vector of PRDM5 gene is successfully constructed,and the stably transfected GV492-PRDM5-Neuro-2a cells are established.

Objective To develop a solvent desorption-gas chromatography method for quantifying malononitrile in workplace air. Methods Malononitrile in workplace air was collected using a silica gel tube and desorbed with methanol. Separation was performed using DB-FFAP capillary column, and detection was performed by hydrogen flame ionization detector. Results The linear ranges of malononitrile were 4.00-600.00 mg/L, with the correlation coefficient of 0.999 92. The detection limit was 0.54 mg/L. The minimum detection concentration was 0.36 mg/m³ (based on a collected air volume of 1.5 L). The average desorption efficiency was 95.0%-101.0%. The within-run and between-run relative standard deviations were 0.9%-5.3% and 1.0%-6.8%, respectively. The sampling efficiency was 100.0%. The samples were stable at 4 ℃ for at least 14 days. Conclusion The solvent desorption-gas chromatography method established in this study for the determination of malononitrile in workplace air is simple to use, with high accuracy and good precision, and is suitable for the determination of malononitrile in workplace air.

OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.

Purpose To investigate the changes in quadriceps femoris microcirculatory perfusion level after low-load blood flow restriction training and its relationship with muscle strength.Materials and Methods Twenty-five healthy subjects were prospectively recruited in the First Affiliated Hospital of Kunming Medical University from September to November 2022.A 200 mmHg pressure cuff was applied at the root of the left thigh for blood flow restriction,and the subjects completed regular knee extension training within 4 weeks.Before the first training session and within 24 hours after the last training session,all subjects underwent scanning with the 3.0T MRI intravoxel incoherent motion sequence and the multi-echo steady-state acquisition three-dimentional imaging sequence.After image post-processing,the quadriceps femoris cross-sectional area,perfusion fraction and pseudo-diffusion coefficient were obtained,and the peak torque was measured using an isokinetic dynamometer.The MRI and muscle strength test parameters before and after training were compared,and correlation analyses were performed between the change of peak torque and the change of perfusion fraction,cross-sectional area,and pseudo-diffusion coefficient respectively.Results After low-load blood flow restriction training,the cross-sectional area of the left quadriceps femoris in subjects increased(t=-4.515,P<0.05).Among its components,the cross-sectional area of the left rectus femoris,vastus intermedius and vastus lateralis all increased(t=-3.302,-2.877,-3.207,all P<0.05).The perfusion fraction value of the left quadriceps femoris increased(t=-5.447,P<0.05);the perfusion fraction values of the left rectus femoris,vastus intermedius,vastus lateralis and vastus medialis all increased(t=-5.723,-4.621,-3.767,-4.682,all P<0.05);the muscle strength of the left quadriceps femoris increased(t=-3.983,P<0.05).There was a highly positive correlation between change of perfusion fraction and peak torque of the left quadriceps femoris in subjects(r=0.708,P<0.05).Conclusion After low-load blood flow restriction training,the changes in quadriceps femoris muscle microperfusion quantified by intravoxel incoherent motion are related to muscle strength,which is helpful for formulating rehabilitation training strategies for young patients.

Purpose To investigate the changes in quadriceps femoris microcirculatory perfusion level after low-load blood flow restriction training and its relationship with muscle strength.Materials and Methods Twenty-five healthy subjects were prospectively recruited in the First Affiliated Hospital of Kunming Medical University from September to November 2022.A 200 mmHg pressure cuff was applied at the root of the left thigh for blood flow restriction,and the subjects completed regular knee extension training within 4 weeks.Before the first training session and within 24 hours after the last training session,all subjects underwent scanning with the 3.0T MRI intravoxel incoherent motion sequence and the multi-echo steady-state acquisition three-dimentional imaging sequence.After image post-processing,the quadriceps femoris cross-sectional area,perfusion fraction and pseudo-diffusion coefficient were obtained,and the peak torque was measured using an isokinetic dynamometer.The MRI and muscle strength test parameters before and after training were compared,and correlation analyses were performed between the change of peak torque and the change of perfusion fraction,cross-sectional area,and pseudo-diffusion coefficient respectively.Results After low-load blood flow restriction training,the cross-sectional area of the left quadriceps femoris in subjects increased(t=-4.515,P<0.05).Among its components,the cross-sectional area of the left rectus femoris,vastus intermedius and vastus lateralis all increased(t=-3.302,-2.877,-3.207,all P<0.05).The perfusion fraction value of the left quadriceps femoris increased(t=-5.447,P<0.05);the perfusion fraction values of the left rectus femoris,vastus intermedius,vastus lateralis and vastus medialis all increased(t=-5.723,-4.621,-3.767,-4.682,all P<0.05);the muscle strength of the left quadriceps femoris increased(t=-3.983,P<0.05).There was a highly positive correlation between change of perfusion fraction and peak torque of the left quadriceps femoris in subjects(r=0.708,P<0.05).Conclusion After low-load blood flow restriction training,the changes in quadriceps femoris muscle microperfusion quantified by intravoxel incoherent motion are related to muscle strength,which is helpful for formulating rehabilitation training strategies for young patients.

OBJECTIVE:Nowadays,machine learning algorithms are gradually being applied to predict stroke and cardiovascular disease.Compared with traditional regression models,machine learning can learn from data to achieve high prediction accuracy by exploring the flexible relationship between a large number of predictive features and outcome variables,providing a new method for the formulation of individualized treatment and rehabilitation programs.This study aims to systematically evaluate stroke functional recovery and prognosis prediction models based on machine learning,comprehensively assessing their predictive performance and clinical application potential to provide references for the development,application,and promotion of related predictive models.METHODS:This review was conducted following the PRISMA(Preferred Reporting Items for Systematic Reviews and Meta-Analyses)guidelines.Relevant literature on stroke prognosis prediction using machine learning methods was selected by searching PubMed,EMbase,Web of Science Core Collection,CNKI,WanFang,and the China Biomedical Literature Database,with the search period from January 1,2014,to July 1,2024.Two researchers independently screened the literature and extracted data based on inclusion and exclusion criteria,using the Prediction model Risk Of Bias ASsessment Tool(PROBAST)to assess model quality.RESULTS:(1)A total of 3 126 articles were obtained in the preliminary search.After screening and exclusion,18 articles were finally included.150 prediction models were constructed using 13 machine learning methods.The three most frequently used methods are Logistic Regression,Random Forest,and Extreme Gradient Boosting(XGBoost).Only one study was externally validated.Eight studies reported how the missing data were handled.(2)In terms of outcome indicators,8 studies used the combination of clinical data and imaging data to build models,9 studies only used clinical data to build models,and 1 study only used imaging data to build models.(3)Each of the 18 studies gave the most important characteristics of the study,with the most mentioned being the National Institute of Health Stroke Scale and age.All studies reported area under curve values ranging from 0.74 to 0.96,with the highest area under curve being 0.96.The overall risk of bias in all models was high.The high risk of bias in the field of model analysis was the main reason for the high risk of overall bias in all models.(4)The results of meta-analysis showed that age and National Institute of Health Stroke Scale score had significant influence on stroke prognosis,with age[MD=8.49,95％CI(6.24,10.75),P＜0.01]and National Institute of Health Stroke Scale score[MD=4.78,95％CI(2.56,7.00),P＜0.01].CONCLUSION:This study systematically evaluated the predictive model of functional recovery and prognosis of stroke based on machine learning,and all the models have good predictive potential.However,future studies should increase the sample size of the included model,adopt prospective studies,and add external validation of the model to improve the stability and prediction accuracy of the model,control the risk of bias,and contribute to the validation and promotion of the model in practical clinical applications.At the same time,the interpolation of missing values is more transparent and accurate.Although existing machine learning models show good predictive performance,it is also important to focus on the functionality and usability of the model,and the inclusion of features will reduce ease of use.We should develop easy to use model interfaces and user-friendly clinical tools to enable medical staff to better apply the model for clinical decision.