OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

Objective To systematically analyze the spatiotemporal distribution characteristics and epidemiological trends of tracheal, bronchus, and lung cancer (TBL) disease burden attributed to air pollution globally and in China and the United States from 1990 to 2021, and to assess the patterns of disease burden changes from 2022 to 2031 based on predictive models, providing a scientific basis for formulating targeted TBL prevention and control strategies. Methods Based on the Global Burden of Disease (GBD) 2021 database, we analyzed the disease burden data of TBL attributed to air pollution globally and in China and the United States from 1990 to 2021. R Studio 4.3.2 software was used to analyze the corresponding trends and the Bayesian age-period-cohort (BAPC) prediction model was used to predict the status of the disease burden of TBL attributed to air pollution in the world and in China and the United States from 2022 to 2031. Results In 2021, China had the highest number of deaths and disability-adjusted life years attributed to air pollution (211 400 patients and 4.8947 million person-years), followed by the United States (6 000 patients and 124 300 person-years). The age-standardized mortality rate (ASMR) and age-standardized disability-adjusted life years rate (ASDR) of TBL due to air pollution in the world and in China and the United States showed a decreasing trend. From 1990 to 2021, the ASMR and ASDR of TBL in China due to air pollution were much higher than those in the United States and the global average. In terms of gender, from 1990 to 2021, the disease burden of male patients with TBL attributed to air pollution was much higher than that of female patients. The BAPC prediction model showed that from 2022 to 2031, the ASMR and ASDR of TBL attributed to air pollution showed an upward trend globally, while they showed a downward trend in China and the United States. Conclusion Over the past 30 years, the air pollution-related TBL disease burden in the world and in China and the United States has continued to decline, but China's disease burden is still significantly higher than the global average. The disease burden in men far exceeds that in women, with men and the population aged ≥50 years being high-risk groups. In the future, the global disease trend may reverse and rise, while China and the United States are expected to continuously decline. However, precise prevention and control for high-risk groups remains a key challenge.

Objective To integrate and analyze the disease burden of esophageal cancer caused by alcohol consumption in China from 1990 to 2019, along with the differences between genders, and predict the trends in disease burden changes from 2020 to 2029 to improve prevention and treatment strategies. Methods The disease burden of esophageal cancer caused by alcohol consumption in China from 1990 to 2019 was extracted and integrated from the 2019 Global Burden of Disease (GBD) database, and the corresponding trend was analyzed using the Joinpoint regression model with Joinpoint 4.9.1.0 software. The gray prediction model [GM (1, 1) ] was used to forecast the disease burden of alcohol-related esophageal cancer in China from 2020 to 2029. Results In 2019, the leading causes of esophageal cancer in China were tobacco, alcohol, high body mass index, and insufficient fruit and vegetable intake, accounting for the first to fifth positions in esophageal cancer deaths. From a gender perspective, in 2019, the death number and standardized mortality rate for males were 18.97 times and 20.00 times higher than for females, respectively. The disability-adjusted life years (DALYs) and standardized DALYs rate for males were 33.08 times and 24.78 times higher than those for females, respectively, indicating a heavier disease burden of alcohol-related esophageal cancer among Chinese males. From 1990 to 2019, the average annual percentage change (AAPC) in deaths and DALYs due to alcohol-related esophageal cancer in China was 2.08% and 1.63%, respectively, showing a continuous upward trend with statistical significance (P<0.05). The AAPC values for standardized mortality rate and standardized DALYs rate from 1990 to 2019 were –0.92% and –1.23%, respectively, showing a continuous downward trend with statistical significance (P<0.05). The population aged ≥55 years was the main group bearing the disease burden among all age groups from 1990 to 2019. The gray prediction model predicted that by 2029, the overall standardized mortality rate and standardized DALYs rate would decrease to 2.94/100 000and 67.94/100 000, with a greater decline in females than in males. Conclusion Over the past 30 years, the disease burden of alcohol-related esophageal cancer in China has slightly decreased. However, the reduction in disease burden is still lower compared to the overall decline in esophageal cancer burden, and the disease burden for males is significantly higher than for females. Focusing on prevention and treatment for males and the elderly population remains a major issue in addressing alcohol-related esophageal cancer in China.

Respiratory syncytial virus(RSV), an RNA virus, is one of the leading pathogens causing respiratory tract infections in infants, the elderly and immunocompromised people, resulting in a huge medical burden of disease worldwide every year. The fusion protein(F protein) on the surface of RSV is the main target of neutralizing antibodies, especially the discovery of its pre-fusion conformation(pre-F), which has laid a key theoretical foundation for the design of a new generation of vaccines. In view of the current limited effective prevention and treatment methods for RSV-related diseases, it is necessary to develop effective vaccines, and vaccines of various technical routes, including subunit vaccines, mRNA vaccines,live attenuated vaccines and viral vector vaccines, are under the development stage. Among them, some RSV vaccine prouducts for the elderly and pregnant women have been approved for marketing. This paper mainly outlines the problems encountered in the early development of RSV vaccines, the current status and progress of RSV vaccine research, in order to provide reference for the follow-up development, clinical evaluation and immunization strategy formulation of RSV vaccines.

Objective To address the personalized differences in motion gait between stroke patients and healthy older adults,as well as the issue of abnormal gait recognition,a deep learning fusion-based approach is proposed to effectively improve the accuracy of abnormal gait recognition.Methods A model fusing convolutional neural networks(CNN)and bidirectional long short-term memory networks(BiLSTM)was adopted,with the introduction of a residual network(ResNet).Unilateral ankle joint movement data at different walking speeds within a comfortable range were collected from healthy older adults and stroke patients.Signals from inertial sensors and electromyography sensors were used as inputs,while gait features were analyzed and gait differences between the two groups were compared.The effectiveness of the model was validated by comparing the classification performance of traditional deep learning models and CNN-ResNet-BiLSTM models with different layer combinations in terms of abnormal gait recognition accuracy.Results The CNN-ResNet-BiLSTM model,which introduced residual connectivity,performed excellently in abnormal gait recognition.Compared with traditional deep learning models such as the gated recurrent unit(GRU)and long short-term memory network(LSTM),its prediction accuracy was improved by 13.6％and 8.36％,respectively.Additionally,compared with other model combinations,this model achieved an overall accuracy of 97.78％.Conclusions The algorithm proposed in this study can be applied to stroke-related abnormal gait detection,providing technique support for the early diagnosis and precise monitoring of such diseases.

Magnetic resonance angiography(MRA)is a non-invasive imaging technique used to observe blood vessels.Quantitative analysis of MRA images enables visualization of vascular pathways,condition,and blood flow dynamics,which is essential for diagnosing vascular diseases such as vascular lesions,stenosis,and occlusions.Vessel segmentation serves as the fundamental basis for quantitative vascular analysis.However,the complex morphology of vessels,difficulties in labeling,and scarcity of accurate 3D vascular annotations pose significant challenges for MRA-based vessel segmentation.A strategy of selectively occluding vessels during model training is proposed to enhance the algorithm's capacity to capture the topological structure of blood vessels,thereby improving the continuity of vessel segmentation results.Additionally,a Refine network is incorporated to refine the binary segmentation results of the segmentation network,thereby further improving segmentation accuracy.Model training and testing are carried out using 42 cases of 3D MRA data from the public MIDAS dataset.For the test set,the 3D U-Net baseline model with vessel occlusion strategy shows a β0 Error of 1.2742±0.2103 and a β1 Error of 0.3393±0.0818,respectively,which are 0.1136 and 0.0280 lower than the baseline.The model integrating vessel occlusion strategy and Refine network achieves an average Dice score of 0.7105±0.0125,which is 0.0028 higher than the baseline.These results demonstrate that the proposed method effectively improves both vascular connectivity and segmentation accuracy.

Objective To evaluate the effects of low-dose intrathecal morphine(ITM)combined with liposomal bupivacaine adductor canal block(LB-ACB)on postoperative analgesia and opioid-sparing efficacy in patients undergoing total knee arthroplasty(TKA).Methods In this randomized,double-blind,controlled trial,80 TKA patients were allocated to either an intervention group(ITM 0.1 mg+LB-ACB,n=40)or a control group(intrathecal saline+LB-ACB,n=40).Primary outcomes included resting/movement visual analog scale(VAS)scores at 6,12,24,48,and 72 hours postoperatively,48-hour morphine consumption,time to first rescue analgesia,and incidence of complications.Results(1)The intervention group showed significantly lower resting and movement VAS scores at 6,12,24,and 48 hours postoperatively compared with controls(all P<0.05),except at 72 hours(P>0.05).(2)The intervention group had a significant reduction in 48-hour morphine consumption(4.58±1.0 mg vs.9.34±4.8 mg,P=0.027),a significantly lower rescue analgesia rate(15.0%vs.47.5%,P=0.002),and a significantly prolonged time to first rescue analgesia(48.8±7.5 h vs.14.5±5.5 h,P<0.001).(3)The intervention group demonstrated a significant decrease in the incidence of nausea(from 15.0%to 35.0%,P=0.039)and vomiting(from 10.0%to 27.5%,P=0.045),but no significant differences were observed in the incidences of pruritus,urinary retention,or motor block(all P>0.05).Conclusion Low-dose ITM(0.1 mg)combined with LB-ACB significantly enhances early postoperative analgesia,reduces opioid consumption,and decreases nausea/vomiting risk,without increasing the risks of other complications.This regimen aligns with enhanced recovery after surgery(ERAS)principles.

Objective To explore the identification method,pathogenesis,clinical characteristics and individualized pharmacotherapy of asymptomatic hyperuricemia after renal transplantation.Methods The pharmacist was on duty at the organ transplant outpatient clinic.During this time,they analyzed and sorted out the medications,identified and differentiated a case of asymptomatic hyperuricemia related to spironolactone in a patient who had undergone a renal transplant,and provided comprehensive care throughout the entire process.Results The asymptomatic hyperuricemia in this patient might be associated with spironolactone,and the adverse reactions of the patient were alleviated by pharmacists through optimizing clinical treatment.Up to now,no hyperuricemia occurred.Conclusions Pharmacists are required to collaborate closely with clinicians to establish medication profiles for patients under long-term follow-up and to closely monitor and evaluate drug-related adverse reactions.Additionally,they should assess the renal function and immune status of transplant recipients promptly and formulate individualized treatment plans in order to enhance the long-term survival of both the transplanted kidneys and the recipients.

Magnetic resonance angiography(MRA)is a non-invasive imaging technique used to observe blood vessels.Quantitative analysis of MRA images enables visualization of vascular pathways,condition,and blood flow dynamics,which is essential for diagnosing vascular diseases such as vascular lesions,stenosis,and occlusions.Vessel segmentation serves as the fundamental basis for quantitative vascular analysis.However,the complex morphology of vessels,difficulties in labeling,and scarcity of accurate 3D vascular annotations pose significant challenges for MRA-based vessel segmentation.A strategy of selectively occluding vessels during model training is proposed to enhance the algorithm's capacity to capture the topological structure of blood vessels,thereby improving the continuity of vessel segmentation results.Additionally,a Refine network is incorporated to refine the binary segmentation results of the segmentation network,thereby further improving segmentation accuracy.Model training and testing are carried out using 42 cases of 3D MRA data from the public MIDAS dataset.For the test set,the 3D U-Net baseline model with vessel occlusion strategy shows a β0 Error of 1.2742±0.2103 and a β1 Error of 0.3393±0.0818,respectively,which are 0.1136 and 0.0280 lower than the baseline.The model integrating vessel occlusion strategy and Refine network achieves an average Dice score of 0.7105±0.0125,which is 0.0028 higher than the baseline.These results demonstrate that the proposed method effectively improves both vascular connectivity and segmentation accuracy.