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.

A 21-year-old unmarried female visited Obstetrics and Gynecology Hospital,Fudan University with the chief complaint,memory decline for more than 20 days and discovery of a pelvic mass for more than 10 days.Through systematic examination,she was diagnosed with teratoma-associated anti-N-methyl-D-aspartate receptor(NMDAR)encephalitis.The patient underwent bilateral ovarian teratoma enucleation,and postoperative pathology revealed immature teratoma grade Ⅰ in the left ovary and mature teratoma in the right ovary.Perioperative combined immunotherapy significantly improved her neuropsychiatric symptoms.The diagnosis and treatment of ovarian teratoma-associated anti-NMDAR encephalitis require multidisciplinary collaboration,with accurate diagnosis,timely surgical resection within 1 month,and early initiation of immunotherapy being critical factors for improving prognosis.Clinicians should remain vigilant to avoid missed or misdiagnosed cases,as this condition can lead to severe outcomes.

We aimed to explore the value of personalized narrative intervention in the nursing care of elderly patients with stroke hemiplegia(SH).A total of 68 elderly patients with SH were selected and divided into two groups using the ball-rolling method,with 34 cases in each group.The control group received routine nursing care,while the observation group received routine care and personalized narrative nursing.The nursing outcomes of the two groups were compared.Compared with the control group,the observation group had lower scores in anxiety and depression(P<0.05),and higher scores in the Functional Evaluation of Cognitive State-Short Screening(FECS-SS)(P<0.05).The rehabilitation effects of normal rate in the observation group was higher than that in the control group(P<0.05).Personalized narrative nursing for SH patients can effectively relieve the patients'negative emotions,enhance their compliance with rehabilitation,and improve the outcomes of rehabilitation.

Objective To propose a multiple regression-variational auto-encoders(MR-VAE)model to realize precise and non-invasive prediction of intra-abdominal pressure(IAP)in critically ill patients.Methods At first,a dataset was constructed by retrospectively analysing baseline characteristics and clinical indicators of 100 critically ill patients admitted to the Intensive Care Unit of Daping Hospital of Army Medical University between 30 August 2019 and 30 March 2021.Then,a MR-VAE prediction model was developed by integrating a feedforward neural network for supervised regression onto a variational autoencoder(VAE)framework and incorporating multiple regression strategies to mitigate feature interference.Finally,the MR-VAE model had its performance evaluated by its comparison with five classical models including support vector machines(SVM),convolutional neural networks(CNN),Scikit-learn integrated model(SIM),multi-layer perceptron(MLP)and K-nearest neighbors(KNN),and its prediction accuracy verified by testing the data of 10 randomly selected patients.Results The MR-VAE model behaved the best when compared with the five classical models,with a mean squared error(MSE)of 0.207,a root mean square error(RMSE)of 0.454,a mean absolute error(MAE)of 0.361,a median absolute deviation(MAD)of 0.243,an explained variance score(EVS)of 0.814 and a R2of 0.823,which also outperformed the five models in fitting performance,convergence and final loss.In random sample testing,the MR-VAE model exhibited high consistency between predicted and actual values.Conclusion The MR-VAE model proposed can accurately predict IAP,which has great potential in reducing the repeated measurements of IAP in critically ill patients and providing new ideas for the early diagnosis and treatment of IAH.

Aim To investigate the effects of Congrong San(CRS)on learning and memory ability,hippocam-pal neuronal injury,and ferroptosis in rats with Alzhei-mer's disease(AD)and to explore the related mecha-nisms.Methods AD rat models were established and divided into Sham,Model,CRS low-dose,CRS medium-dose,CRS high-dose,and memantine groups.After treatment,Morris water maze,HE and Nissl staining,transmission electron microscopy,immunofluorescence staining,Western blot,and kit assays were performed to assess learning and memory ability,hippocampal neuro-nal injury,ferroptosis-related indicatorsand glucose reg-ulated protein 78 ku(GRP78)-(proteinkinaseR-li-keERkinase)PERK-(activating transcription factor 4)ATF4 pathway protein expression.Results Com-pared with the model group,rats in the CRS medium-and high-dose groups and the memantine group showed significant improvement in learning and memory abili-ty,reduced hippocampal neuronal injury,increased number of Nissl bodies,and ameliorated endoplasmic reticulum swelling and mitochondrial damage.In addi-tion,the expressions of GRP78,p-PERK/PERK,and ATF4 were downregulated,while GPX4 expression was upregulated in the CRS medium-and high-dose groups and the memantine group.Moreover,MDA content de-creased,and SOD and GSH-PX levels increased in these groups.Conclusions CRS can improve the learning and memory ability in AD rats,reduce hipp-ocampal neuronal injury and ferroptosis,and its mecha-nism may be related to the inhibition of the GRP78-PERK-ATF4 pathway,enhancement of GPX4 expres-sion,and reduction of oxidative stress levels,providing a new approach for the clinical treatment of AD.

Aim To investigate the protective effect of Gualou Guizhi granules(GLGZG)on neuronal injury induced by LPS-activated microglia based on Notch signaling pathway.Methods LPS-activated microglia were co-cultured with neurons to construct neuron inju-ry models,and the cells were divided into the control group,model group,Notch inhibitor(DAPT)group,GLGZG(50,100,200 mg·L-1)group,DAPT+100 mg·L-1GLGZG group.After intervention,the activity of HT22 cells was detected by CCK-8 method,and rel-ative mRNA expression was detected by real-time PCR.The relative protein expression was detected by Western blot.Results Compared with the model group,after GLGZG intervention,the cell activity was significantly improved,GLGZG decreased IL-6,IL-12,Bax,Notch 1,caspase-3,Delta-1,NICD,RBPSUH,HES1 expression,and increased Bcl-2 expression(P＜0.05).Compared with the model group,the NICD,RBPSUH and HES1 mRNA and protein expressions significantly decreased after DAPT treatment(P＜0.05),and there was no superposition effect with GLG-ZG.Conclusion GLGZG may play a neuroprotective role by inhibiting inflammatory factors and apoptosis,and inhibiting Notch signaling pathway.

Objective:To compare the clinical efficacy of Minimally Invasive Surgery (MIS) and traditional Posterior Spinal Fusion (PSF) in treating children with paralytic scoliosis with pelvic obliquity (PSPO) following spinal cord injury.Methods:A retrospective analysis was conducted on the data of 25 patients with PSPO who underwent surgical treatment at the Drum Tower Hospital affiliated with Nanjing University Medical School from January 2017 to June 2023. The cohort included 4 males and 21 females, aged 12.3±2.8 years (range 9-14 years). Patients were divided into the MIS group (12 cases) and the PSF group (13 cases). Radiological parameters were measured preoperatively, postoperatively, and at the last follow-up. Surgical time, intraoperative blood loss, intraoperative blood transfusion volume, length of hospital stay, total hospitalization costs, and complications were recorded. The Scoliosis Research Society questionnaires-22 (SRS-22) Chinese version were used to assess patient satisfaction and efficacy.Results:There were no statistically significant differences between the MIS and PSF groups in age, gender, Risser sign, preoperative Cobb angle for scoliosis, pelvic tilt angle, or local kyphosis angle ( P>0.05). The MIS group demonstrated surgical time of 176±30 minutes, intraoperative blood loss of 300±70 ml, blood transfusion volume of 280±175 ml, and total hospitalization costs of 87'800± 13'300 yuan, all of which were lower than PSF group, with values of 280±91 minutes, 1'433±116 ml, 1'351±996 ml, and 14'8400±26'100 yuan, respectively. These differences were statistically significant ( t=3.789, P=0.001; t=29.328, P<0.001; t=3.667, P=0.001; t=7.271, P<0.001). In the MIS group, preoperative, postoperative, and last follow-up Cobb angles were 79.11°±6.74°, 35.86°±4.98°, and 36.27°±4.84° respectively; pelvic tilt angles were 24.79°±5.58°, 9.18°±3.32°, and 8.79°±2.94°; local kyphosis angles were 38.84°±4.18°, 12.96°±4.87°, and 11.43°±6.08°, respectively. Postoperative and last follow-up angles were significantly reduced compared to preoperative values, with statistically significant differences ( P<0.05). In the PSF group, preoperative, postoperative, and last follow-up Cobb angles were 82.06°±9.26°, 34.75°±5.14°, and 35.15°±5.04° respectively; pelvic tilt angles were 26.60°±6.21°, 10.12°±3.21°, and 9.91°±2.97°; local kyphosis angles were 40.92°±7.04°, 10.92°±7.26°, and 14.02°±5.58°, respectively. Differences from preoperative to postoperative measurements were statistically significant ( P<0.05). At the last follow-up, both groups showed no significant loss of scoliosis correction, and there were no statistically significant differences between the groups postoperatively or at the last follow-up ( P>0.05). In the MIS group, one case of superficial surgical site infection and one case of postoperative atelectasis occurred. In the PSF group, two cases of deep surgical site infection, one case of poor screw placement, and two cases were transferred to the ICU postoperatively due to excessive intraoperative bleeding. Preoperative SRS-22 total scores were 2.0±0.6 for PSF and 2.1±0.4 for MIS. Postoperative SRS-22 total scores (excluding satisfaction) were 3.0±0.5 for PSF and 2.9±0.3 for MIS. The within-group differences from preoperative to postoperative were statistically significant ( P<0.05), while the between-group differences from preoperative to postoperative were not statistically significant ( P>0.05). Conclusion:Compared to the PSF technique, MIS can shorten surgery time, reduce intraoperative blood loss and perioperative complications, and decrease hospitalization costs. MIS can achieve similar early clinical efficacy.

Objective To investigate the effect of different residence time of povidone iodine on the disinfection effect of conjunctival sac in senile cataract surgery.Methods A total of 396 senile patients who underwent cataract surgery in our hospital were selected and randomly divided into the long-residence group and the short-residence group,with 198 cases in each group.The patients of the long-residence group and the short-residence group were disinfected with a disposable filling 5%povidone iodine on conjunctival sac for 3 minutes and 30 seconds before surgery,respectively.The bacterial detection and bacterial distribution on conjunctival sac,and corneal epithelial cell damage were compared between the two groups.The subjective comfort conditions such as ocular itching,foreign body sensation,erythema,and photophobia after disinfection were recorded.Results There was no significant difference in the bacterial detection rate before and after disinfection between the two groups(P>0.05).The most frequently detected bacteria in the short-residence group was Staphylococcus epidermidis,accounting for 28.79%and 33.33%before and after disinfection,respectively.The incidences of ocular itching,foreign body sensation,erythema,and photophobia after disinfection in the short-residence group were significantly lower than those in the long-residence group(P<0.05),while the rate of corneal epithelial cell damage in the long-residence group was significantly higher than that in the short-residence group(P<0.05).Conclusion The use of disposable filling 5%povidone iodine for disinfection on conjunctival sac for 30 seconds before surgery can achieve a comparable disinfection effect to disinfection for 3 minutes,which has less effect on corneal epithelial cells,and higher ocular comfort.

Objective:To evaluate the relationship between postoperative changes in femoral head coverage (FHC) after S 2-Alar-Iliac (S 2AI) screw fixation and the development of sagittal imbalance during follow-up in patients with adult spinal deformity (ASD), providing insights for clinical assessment and treatment strategies. Methods:A consecutive cohort of 98 ASD patients who underwent S2AI fixation between September 2019 and September 2021 was retrospectively analyzed. Patients were divided into two groups based on changes in femoral head coverage (ΔFHC): the FHC-C group (upper quartile ΔFHC, 25 cases) and the FHC-NC group (lower quartile ΔFHC, 24 cases). Additionally, patients were classified into proximal junctional kyphosis (PJK) and non-PJK groups based on their clinical outcomes at the last follow-up. Standing full-spine anteroposterior and lateral X-rays were taken preoperatively, postoperatively, and at the two-year follow-up to measure and document the following spinal parameters: Cobb angle, proximal lumbar lordosis (PLL), distal lumbar lordosis (DLL), lumbar lordosis (LL), lordosis distribution index (LDI), sagittal vertical axis (SVA), coronal balance distance (CBD), thoracic kyphosis (TK), T 1 pelvic angle (T 1PA), pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), PI-LL, and proximal junctional angle (PJA). Parameters related to hip joint coverage included: femoral head coverage (FHC), lateral center-edge angle (LCE angle), acetabular index (AI), Sharp angle, and extrusion index (EI). Comparisons of radiographic indicators between the two groups were performed at preoperative, postoperative, and final follow-up assessments. The visual analogue scale (VAS) was used to evaluate the hip pain and back pain. Results:At final follow-up, the incidence of PJK was significantly higher in the FHC-NC group [37.5% (9/24)] compared to the FHC-C group [16.0% (4/25)] (χ 2=3.952, P=0.042). Moreover, the increase in sagittal vertical axis (ΔSVA) was significantly greater in the FHC-NC group (35.9±44.7 mm vs. 14.6±31.8 mm, t=2.216, P=0.031). Patients with PJK had significantly higher preoperative T 1PA (36.8°±10.8° vs. 31.9°±18.4°, t=2.150, P=0.034) and lower immediate postoperative ΔFHC (1.7%±1.5% vs. 3.3%±2.5%, t=2.987, P=0.004), as well as lower changes in lateral center-edge angle during follow-up (0.3°±3.0° vs. 1.1°±8.9°, t=2.334, P=0.022). Pearson correlation analysis revealed significant negative correlations between postoperative ΔFHC and both ΔSVA ( r=-0.374, P=0.008) and proximal junctional angle changes (ΔPJA, r=-0.429, P=0.006). Additionally, increases in VAS leg pain scores correlated negatively with immediate postoperative FHC ( r=-0.314, P=0.025) and ΔFHC ( r=-0.298, P=0.031). Logistic regression indicated that immediate postoperative ΔFHC was a protective factor against PJK [ OR=0.722, 95% CI (0.541, 0.963), P=0.009), with a ROC-determined optimal ΔFHC cut-off of 3.90% (AUC=0.723, Youden index=0.847). Conclusions:Postoperative evaluation of femoral head coverage is clinically important for ASD patients undergoing S2AI screw fixation. A pre-to-post ΔFHC below 3.90% may indicate reduced hip compensation capacity, increasing risks for hip pain, sagittal imbalance progression, and PJK postoperatively.