Objective To explore the effect of remote ischemic preconditioning (RIPC) on preoperative heart rate variability in patients with heart valves. Methods Patients scheduled to undergo on-pump cardiac valve surgery in the Department of Cardiovascular Surgery, General Hospital of Northern Theater Command, between January and July 2022 were initially enrolled. Eligible patients were randomly assigned at a 1 : 1 ratio to either the RIPC group or the control group. Relevant indicators of heart rate variability [standard deviation of NN interval (SDNN), standard deviation of mean value of NN interval in every five minutes (SDANN), mean square root of difference between consecutive NN intervals (RMSSD), percentage of adjacent RR interval>50 ms (PNN50), low frequency (LF) component, high frequency (HF) component and LF/HF] at 8 hours in the morning on the surgical day between two groups were compared. Results A total of 118 patients were initially assessed. After screening, 58 patients were excluded, and 60 patients provided written informed consent and were enrolled in the trial, with 30 allocated to the RIPC group and 30 to the control group. Seven patients in the control group and 5 patients in the RIPC group were subsequently excluded due to missing heart rate variability data resulting from cancelled operations. Finally, 23 patients in the control group and 25 patients in the RIPC group were included in the analysis. There was no statistical difference in baseline characteristics between the two groups, and there was no significant difference in heart rate variability 24 hours before intervention (P>0.05). After the intervention measures were taken, the comparison of the results of heart rate variability at 8 hours on the day of operation showed that SDNN and SDANN of patients in the RIPC group were higher than those in the control group, with statistical differences (P<0.05). Conclusion RIPC can stabilize the preoperative heart rate variability of patients undergoing cardiac valve surgery.

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.

Cardiovascular diseases and psychological disorders represent two major threats to human physical and mental health. Research on electrocardiogram (ECG) signals offers valuable opportunities to address these issues. However, existing methods are constrained by limitations in understanding ECG features and transferring knowledge across tasks. To address these challenges, this study developed a multi-resolution feature encoding network based on residual networks, which effectively extracted local morphological features and global rhythm features of ECG signals, thereby enhancing feature representation. Furthermore, a model compression-based continual learning method was proposed, enabling the structured transfer of knowledge from simpler tasks to more complex ones, resulting in improved performance in downstream tasks. The multi-resolution learning model demonstrated superior or comparable performance to state-of-the-art algorithms across five datasets, including tasks such as ECG QRS complex detection, arrhythmia classification, and emotion classification. The continual learning method achieved significant improvements over conventional training approaches in cross-domain, cross-task, and incremental data scenarios. These results highlight the potential of the proposed method for effective cross-task knowledge transfer in ECG analysis and offer a new perspective for multi-task learning using ECG signals.
Humans ; Electrocardiography/methods* ; Mental Health ; Algorithms ; Signal Processing, Computer-Assisted ; Machine Learning ; Arrhythmias, Cardiac/diagnosis* ; Cardiovascular Diseases ; Neural Networks, Computer ; Mental Disorders

BACKGROUND: Lung cancer is one of the most common malignant tumors worldwide and a major cause of cancer-related deaths. Early-stage lung cancer is often manifested as pulmonary nodules, and accurate assessment of the malignancy risk is crucial for prolonging survival and avoiding overtreatment. This study aims to construct a model based on image feature parameters automatically extracted by artificial intelligence (AI) to evaluate its effectiveness in predicting the malignancy of part-solid nodule (PSN). METHODS: This retrospective study analyzed 229 PSN from 222 patients who underwent pulmonary nodule resection at Lanzhou University Second Hospital between October 2020 and February 2025. According to pathological results, 45 cases of benign lesions and precursor glandular lesion were categorized into the non-malignant group, and 184 cases of pulmonary malignancies were categorized into the malignant group. All patients underwent preoperative chest computed tomography (CT), and AI software was used to extract imaging feature parameters. Univariate analysis was used to screen significant variables; variance inflation factor (VIF) was calculated to exclude highly collinear variables, and LASSO regression was further applied to identify key features. Multivariate Logistic regression was used to determine independent risk factors. Based on the selected variables, five models were constructed: Logistic regression, random forest, XGBoost, LightGBM, and support vector machine (SVM). Receiver operating characteristic (ROC) curves were used to assess the performance of the models. RESULTS: The independent risk factors for the malignancy of PSN include roughness (ngtdm), dependence variance (gldm), and short run low gray-level emphasis (glrlm). Logistic regression achieved area under the curves ( AUCs) of 0.86 and 0.89 in the training and testing sets, respectively, showing good performance. XGBoost had AUCs of 0.78 and 0.77, respectively, demonstrating relatively balanced performance, but with lower accuracy. SVM showed an AUC of 0.93 in the training set, which decreased to 0.80 in the testing set, indicating overfitting. LightGBM performed excellently in the training set with an AUC of 0.94, but its performance declined in the testing set, with an AUC of 0.88. In contrast, random forest demonstrated stable performance in both the training and testing sets, with AUCs of 0.89 and 0.91, respectively, exhibiting high stability and excellent generalizability. CONCLUSIONS The random forest model constructed based on independent risk factors demonstrated the best performance in predicting the malignancy of PSN and could provide effective auxiliary predictions for clinicians, supporting individualized treatment decisions.
.
Humans ; Male ; Female ; Lung Neoplasms/pathology* ; Middle Aged ; Retrospective Studies ; Artificial Intelligence ; Aged ; Tomography, X-Ray Computed ; Adult ; Solitary Pulmonary Nodule/diagnostic imaging* ; ROC Curve

ObjectiveTo observe the effects of Hei Xiaoyaosan on the learning and memory abilities of Alzheimer's disease model mice （APP/PS1 mice）， and to explore its mechanism through the inflammatory cascade mediated by nucleotide-binding oligomerization domain-like receptor family pyrin domain-containing 3 （NLRP3）/cysteine aspartate-specific protease （Caspase-1）/gasdermin D （GSDMD） signaling pathway. MethodsSPF-grade 4-month-old APP/PS1 mice were randomly divided into the model group， MCC950 group， and Hei Xiaoyaosan high-， medium-， and low-dose groups. C57BL/6J mice were used as the blank group. After 7 days of adaptive feeding， mice in each group were intervened. The Hei Xiaoyaosan high-， medium-， and low-dose groups were given corresponding doses by gavage （25.79， 12.90， 6.45 g·kg^-1·d^-1）， the MCC950 group was intraperitoneally injected with 10 mg·kg^-1·2 d^-1， and the blank group received the same volume of physiological saline by gavage. After 90 days of intervention， the learning and memory abilities were assessed using the Y maze and Morris water maze tests. The structural changes of hippocampal neurons were observed by hematoxylin-eosin （HE） staining. The expression of amyloid precursor protein （APP） in the hippocampal CA3 region was detected by immunohistochemistry. Enzyme-linked immunosorbent assay （ELISA） was used to measure the levels of interleukin （IL）-10， IL-18， and IL-1β in the hippocampus. Western blot was applied to detect the protein expression of NLRP3， Caspase-1， GSDMD， and GSDMD-N in the hippocampus. Immunofluorescence was used to detect the co-localization of GSDMD-N and ionized calcium-binding adapter molecule-1 （Iba-1） in the hippocampus. Results① In the Y maze test， compared with the blank group， the spontaneous alternation rate of the model group was significantly reduced （P<0.01）. Compared with the model group， the spontaneous alternation rate in the Hei Xiaoyaosan high- and low-dose groups was significantly increased （P<0.01）. ② In the Morris water maze test， during the 1-4 days of the location navigation test， the escape latency time of mice decreased with the extension of training time. On day 4， compared with the blank group， the model group showed a significantly increased escape latency （P<0.05）. Compared with the model group， the MCC950 group and the Hei Xiaoyaosan low-dose group showed significantly reduced escape latency （P<0.05）. In the spatial exploration experiment， compared with the blank group， the number of platform crossings in the model group was significantly reduced （P<0.01）. Compared with the model group， the Hei Xiaoyaosan low-dose group showed significantly increased platform crossings （P<0.05）. ③ HE staining showed that， compared with the blank group， the hippocampal CA3 cells of the model group were damaged， arranged loosely and irregularly， swollen， with unclear boundaries， and the nuclei were pyknotic and deeply stained. MCC950 and all doses of Hei Xiaoyaosan improved the hippocampal CA3 cell damage in APP/PS1 mice to varying degrees. ④ Immunohistochemical results indicated that， compared with the blank group， the expression of APP in the hippocampal CA3 region was significantly increased in the model group （P<0.01）. MCC950 and all doses of Hei Xiaoyaosan could reduce the expression of APP in the hippocampal CA3 region of APP/PS1 mice （P<0.01）. ⑤ ELISA results showed that the levels of IL-18 and IL-1β in the hippocampus of mice in the model group were significantly increased， and IL-10 levels were significantly reduced （P<0.01）. Compared with the model group， the IL-18 levels in the MCC950 group and the Hei Xiaoyaosan medium- and low-dose groups were significantly reduced （P<0.01）. IL-1β levels in the hippocampus of the MCC950 group and Hei Xiaoyaosan high-， medium-， and low-dose groups were significantly decreased （P<0.01）. The IL-10 levels in the hippocampus of the MCC950 group and the Hei Xiaoyaosan medium- and low-dose groups were increased （P<0.05， P<0.01）. ⑥ Western blot results showed that compared with the blank group， the protein levels of NLRP3， Caspase-1， GSDMD， and GSDMD-N in the hippocampus of the model group were significantly elevated （P<0.01）. Compared with the model group， the content of NLRP3 and Caspase-1 in the hippocampus of the treated groups was decreased （P<0.05， P<0.01）. The content of GSDMD in the hippocampus of the Hei Xiaoyaosan high-， medium-， and low-dose groups was reduced （P<0.05， P<0.01）， and the content of GSDMD-N in the hippocampus of the Hei Xiaoyaosan medium- and low-dose groups was decreased （P<0.05， P<0.01）. ⑦ Immunofluorescence results showed that， compared with the blank group， the co-expression of GSDMD-N and Iba-1 in the hippocampus of the model group was significantly increased （P<0.01）. Compared with the model group， the co-expression of GSDMD-N and Iba-1 in the treated groups was significantly reduced （P<0.01）. ConclusionHei Xiaoyaosan may regulate the NLRP3/Caspase-1/GSDMD signaling pathway to affect the release of inflammatory factors， alleviate neuroinflammation，improve hippocampal histopathological changes，and improve learning and memory deficits，thus providing potential therapeutic benefits for Alzheimer's disease.

Cerebrovascular diseases, characterized by high incidence, disability, and mortality rates, have emerged as a leading global cause of death and long-term disability. Organoid technology, a three-dimensional in vitro culture system derived from stem cells or tissue cells, enables the simulation of organ development, physiological processes, and pathological mechanisms, demonstrating significant potential in cerebrovascular disease research and therapeutic development. This review summarizes recent advances in organoid applications for cerebrovascular diseases, with a focus on strategies for constructing vascularized cerebral organoids, including in vivo transplantation, in vitro culture systems, and bioengineering approaches. Studies reveal that these models not only recapitulate neurovascular unit interactions but also serve as powerful platforms for drug screening and mechanistic investigations, offering novel therapeutic strategies for cerebrovascular disorders. Current challenges include insufficient vascularization efficiency and limited integration capacity with host tissues. Future integration of gene editing, microfluidic chips, and high-throughput 3D bioprinting technologies is expected to enhance the functionality and clinical translatability of vascularized cerebral organoids, thereby advancing personalized medicine and precision healthcare.

Objective To compare the mid- and long-term clinical results of mitral valve plasty (MVP) and mitral valve replacement (MVR) in the treatment of functional mitral regurgitation (FMR). Methods Patients with FMR who underwent surgical treatment in the Department of Cardiovascular Surgery of the General Hospital of Northern Theater Command from 2012 to 2021 were collected. The patients who underwent MVP were divided into a MVP group, and those who underwent MVR into a MVR group. The clinical data and mid-term follow-up efficacy of two groups were compared. Results Finally 236 patients were included. There were 100 patients in the MVP group, including 53 males and 47 females, with an average age of (61.80±8.03) years. There were 136 patients in the MVR group, including 72 males and 64 females, with an average age of (61.29±8.97) years. There was no statistical difference in baseline data between the two groups (P＞0.05). There was no statistical difference between the two groups in the extracorporeal circulation time, aortic occlusion time, postoperative hospital and ICU stay, intraoperative blood loss, or hospitalization death (P＞0.05), but the time of mechanical ventilation in the MVP group was significantly shorter than that in the MVR group (P=0.022). The total follow-up rate was 100.0%, the longest follow-up was 10 years, and the average follow-up time was (3.60±2.55) years. There were statistical differences in the left atrial diameter, left ventricular end-diastolic diameter, left ventricular end-systolic diameter and cardiac function between the two groups compared with those before surgery (P<0.05). The postoperative left ventricular ejection fraction in the MVP group was statistically higher than that before surgery (P=0.002), but there was no statistical difference in the MVR group before and after surgery (P=0.658). The left atrial diameter in the MVP group was reduced compared with the MVR group (P=0.026). The recurrence rate of mitral regurgitation in the MVP group was higher than that in the MVR group, and the difference was statistically significant (10.0% vs. 1.5%, P=0.003). There were 14 deaths in the MVP group and 19 in the MVR group. The cumulative survival rate (P=0.605) and cardiovascular events-free survival rate (P=0.875) were not statistically significant between the two groups by Kaplan-Meier survival analysis. Conclusion The safety, and mid- and long-term clinical efficacy of MVP in the treatment of FMR patients are better than MVR, and the left atrial and left ventricular diameters are statistically reduced, and cardiac function is statistically improved. However, the surgeon needs to be well aware of the indications for the MVP procedure to reduce the rate of mitral regurgitation recurrence.

Objective:To investigate the factors influencing the prognosis of critically ill patients and the predictive value of the C-reactive protein-albumin-lymphocyte (CALLY) index.Methods:A retrospective analysis was conducted on the clinical data of 122 critically ill patients admitted to Guoyao Dongfeng General Hospital affiliated with Hubei University of Medicine from June 2022 to December 2023. Patients were divided into a death group and a survival group based on their 28-day prognosis. Clinical data were compared between the two groups to analyze the factors influencing prognosis and assess the predictive value of various indicators. Normally distributed measurement data were expressed as Mean±SD, and intergroup comparisons were performed by independent samples t-test; non-normally distributed measurement data were expressed as M( Q1,Q3), and intergroup comparisons were performed by the Mann-Whitney U test. Counting data were expressed as case (%), and intergroup comparisons were performed by the χ2 test. Multivariate logistic regression was used to analyze factors influencing patient prognosis, and receiver operating characteristic (ROC) curves were plotted to analyze the predictive value of each indicator. Results:The Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score, blood lactate, CRP, and B-type natriuretic peptide levels in the death group were higher than those in the survival group [22 (17, 30) points vs. 17 (14, 22) points, (4.8±1.4) mmol/L vs. (3.3±1.0) mmol/L, 134 (83, 2 381) mg/L vs. 13 (10, 27) mg/L, 259 (111, 592) ng/L vs. 108 (40, 247) ng/L; Z=3.04, P=0.002; t=5.79, P<0.001; Z=8.57, P<0.001; Z=3.28, P=0.001, respectively]. Albumin, neutrophil count, lymphocyte count (LYC), and the CALLY index were lower in the death group than in the survival group [(31±5) g/L vs. (37±6) g/L, (58±9)×10 9/L vs. (63±10)×10 9/L, 0.6 (0.4, 0.8)×10 9/L vs. 1.3 (0.8, 1.7)×10 9/L, 0.03 (0.02, 0.11) vs. 0.26 (0.13, 0.49); t=6.05, P<0.001; t=3.04, P=0.003; Z=5.82, P<0.001; Z=6.52, P<0.001, respectively]. Multivariate logistic regression analysis indicated that the APACHE Ⅱ score and CRP were risk factors for poor prognosis in critically ill patients ( OR=1.349, 95% CI: 1.004-1.821, P=0.048; OR=1.006, 95% CI: 1.003-1.010, P=0.001, respectively), while LYC and the CALLY index were protective factors ( OR=0.297, 95% CI: 0.111-0.795, P=0.016; OR=0.989, 95% CI: 0.955-0.999, P=0.001, respectively). The area under the ROC curve for the CALLY index predicting 28-day mortality in critically ill patients was 0.872 (95% CI: 0.800-0.926), which was higher than that of the APACHE Ⅱ score, LYC, and CRP [0.673 (95% CI: 0.582-0.756), 0.664 (95% CI: 0.573-0.748), 0.576 (95% CI: 0.482-0.665), respectively]. The cut-off values were 0.06, 20 points, 0.8×10 9/L, and 50 mg/L, respectively. When the CALLY index was 0.06, the specificity was 97.65%, the sensitivity was 72.97%, and the Youden index was 0.706. Conclusions:The APACHE Ⅱ score, CRP, LYC, and CALLY index are all factors influencing the prognosis of critically ill patients. The CALLY index has certain predictive value, but its false negative rate is relatively high. Further combination with other indicators is needed to improve its predictive value.

Objective:To explore the correlation between the oxygenation index and the monitored parameters of lung injury view in ventilator,so as to provide a basis for monitoring the change of oxygenation index by the parameters of lung injury view.Methods:A total of 28 patients who received mechanical ventilation treatment and admitted to respiratory department of the First Affiliated Hospital with Nanjing Medical University from May 14 to September 30,2024 were selected.The monitored parameters of lung injury views of them during mechanical ventilation treatment with invasive ventilators were collected,including platform pressure(Pplat),driving pressure(ΔP),mechanical energy(Emech),ratio of tidal volume to ideal body weight(TVe/IBW),etc..Then,the data of oxygenation index were obtained.Finally,135 valid data were screened out.The 84 pieces of data with oxygenation index≤300mmHg were included in low oxygen group,and 51 data with oxygenation index>300mmHg were included in normal group.The correlations between oxygenation index and Pplat,ΔP,Emech,TVe/IBW were conducted by pairwise analysis.Results:In the low oxygen group,the oxygenation index was[(196.302±74.522)mmHg]significantly negative correlation with Emech[(12.686±5.401)J/min](ρ=-0.425,P<0.05),but not was significant correlation with Pplat,ΔP,and TVe/IBW(P>0.05).In the normal group,there were not significant correlations between the oxygenation index and Pplat,ΔP,Emech and TVe/IBW(P>0.05).There was a significant negative correlation between Emech and rank of oxygenation index(β=-0.389,P<0.05)in the low oxygen group.Conclusion:The visualization function of the Emech of the monitored parameters in lung injury view of ventilator can provide help for monitoring the changes of oxygenation index of patients in the low oxygen group,and can improve the oxygenation of patients by optimizing mechanical energy.However,it is necessary to balance the contradiction between ventilation intensity and protection for lung.