AIM: To explore the accuracy and consistency of predicting postoperative vault in implantable collamer lens(ICL)surgery based on the NK and KS formulas of CASIA2 OCT.METHODS: Retrospective case study. The study included 85 patients(170 eyes)who underwent ICL implantation for the correction of moderate to high myopia at the Refractive Surgery Center of Lunan Eye Hospital from October 2022 to October 2023. Routine examinations including ultrasound biomicroscopy(UBM), corneal endothelium assessment, optometry, intraocular pressure measurement, fundus examination, and corneal topography were performed. Patients who met the indications for ICL surgery were selected, and the recommended lens size was determined based on parameters such as corneal diameter and anterior chamber depth using standard methods. A CASIA2 OCT examination was used to obtain the corresponding lens size and predicted vault using the NK and KS formulas. The differences and consistency between the predicted vault by NK and KS formulas and the actual vault at 1 wk, 1, 3, and 6 mo postoperatively were analyzed. Paired t-tests were used to analyze the differences between the predicted vault by NK and KS formulas and the actual vault at each postoperative time point. The consistency between the predicted vault by NK and KS formulas and the actual vault at each time point was assessed using Bland-Altman analysis. Multivariate linear regression was used to analyze the correlation between actual vault and various anterior segment parameters.RESULTS: The NK formula showed statistically significant differences between the predicted values and the actual vault at 1, 3, and 6 mo postoperatively(all P<0.05), but no statistically significant difference was observed compared to the actual vault at 1 wk postoperatively(P=0.340). The KS formula demonstrated a statistically significant difference between the predicted values and the actual vault measured at 6 mo postoperatively(P<0.05). The lens vector height and lens model were influcing facotrs for the vault at 6 mo postoperatively(all P<0.05). Both the NK formula and KS formula had the highest consistency between their predicted vault and the actual vault at 6 mo postoperatively.CONCLUSION: The NK and KS formulas of CASIA2 OCT showed the highest consistency between the predicted postoperative vault and the actual vault at 6 mo postoperatively. Both formulas can serve as effective references for ICL model selection.

Cancer is the result of evolving crosstalk between neoplastic cell and its immune microenvironment. In recent years, immune therapeutics targeting T lymphocytes, such as immune checkpoint blockade (ICB) and CAR-T, have made significant progress in cancer treatment and validated targeting immune cells as a promising approach to fight human cancers. However, responsiveness to the current immune therapeutic agents is limited to only a small proportion of solid cancer patients. As major components of most solid tumors, myeloid cells played critical roles in regulating the initiation and sustentation of adaptive immunity, thus determining tumor progression as well as therapeutic responses. In this review, we discuss emerging data on the diverse functions of myeloid cells in tumor progression through their direct effects or interactions with other immune cells. We explain how different metabolic reprogramming impacts the characteristics and functions of tumor myeloid cells, and discuss recent progress in revealing different mechanisms-chemotaxis, proliferation, survival, and alternative sources-involved in the infiltration and accumulation of myeloid cells within tumors. Further understanding of the function and regulation of myeloid cells is important for the development of novel strategies for therapeutic exploitation in cancer.
Humans ; Tumor Microenvironment/immunology* ; Myeloid Cells/immunology* ; Neoplasms/therapy* ; Animals

OBJECTIVE: To explore the feasibility of incorporating simple bedside indicators into death predictive model for elderly critically ill patients based on interpretability machine learning algorithms, providing a new scheme for clinical disease assessment. METHODS: Elderly critically ill patients aged ≥ 65 years who were hospitalized in the intensive care unit (ICU) of Tacheng People's Hospital of Ili Kazak Autonomous Prefecture from June 2017 to May 2020 were retrospectively selected. Basic parameters including demographic characteristics, basic vital signs and fluid intake and output within 24 hours after admission, as well acute physiology and chronic health evaluation II (APACHE II), Glasgow coma score (GCS) and sequential organ failure assessment (SOFA) were also collected. According to outcomes in hospital, patients were divided into survival group and death group. Four datasets were constructed respectively, namely baseline dataset (B), including age, body temperature, heart rate, pulse oxygen saturation, respiratory rate, mean arterial pressure, urine output volume, infusion volume, and crystal solution volume; B+APACHE II dataset (BA), B+GCS dataset (BG), and B+SOFA dataset (BS). Then three machine learning algorithms, Logistic regression (LR), extreme gradient boosting (XGboost) and gradient boosting decision tree (GBDT) were used to develop the corresponding mortality predictive models within four datasets. The feature importance histogram of each prediction model was drawn by SHapley additive explanation (SHAP) method. The area under curve (AUC), accuracy and F1 score of each model were compared to determine the optimal prediction model and then illuminate the nomogram. RESULTS: A total of 392 patients were collected, including 341 in the survival group and 51 in the death group. There were statistically significant differences in heart rate, pulse oxygen saturation, mean arterial pressure, infusion volume, crystal solution volume, and etiological distribution between the two groups. The top three causes of death were shock, cerebral hemorrhage, and chronic obstructive pulmonary disease. Among the 12 prognostic models trained by three machine learning algorithms, overall performance of prognostic models based on B dataset was behind, whereas the LR model trained by BA dataset achieved the best performance than others with AUC of 0.767 [95% confidence interval (95%CI) was 0.692-0.836], accuracy of 0.875 (95%CI was 0.837-0.903) and F1 score of 0.190. The top 3 variables in this model were crystal solution volume with first 24 hours, heart rate and mean arterial pressure. The nomogram of the model showed that the total score between 150 and 230 were advisable. CONCLUSION The interpretable machine learning model including simple bedside parameters combined with APACHE II score could effectively identify the risk of death in elderly patients with critically illness.
Humans ; Critical Illness ; Machine Learning ; Aged ; Algorithms ; Intensive Care Units ; Retrospective Studies ; APACHE ; Prognosis ; Organ Dysfunction Scores ; Hospital Mortality ; Male ; Female

AIM: To evaluate the changes in corneal epithelial thickness(CET)and corneal optical density(CD)after smart pulse technology(SPT)-assisted transepithelial photorefractive keratectomy(TPRK)and analyze their correlation.METHODS: The prospective study included 60 patients(120 eyes)with myopia and myopic astigmatism who underwent SPT-TPRK in the ophthalmology department at the First Affiliated Hospital of Xinxiang Medical University between February and August 2023. Changes in CET and CD were evaluated preoperatively and at 1 wk, 1 and 3 mo postoperatively.RESULTS: A total of 14 cases(28 eyes)were lost to follow-up, and 3 patients(6 eyes)with postoperative haze were excluded from this study, resulting in a final inclusion of 43 patients(86 eyes). At 1 wk after SPT-TPRK, CET had statistically significantly thickened compared to preoperative levels(P<0.05), particularly in the CET at 0-2 mm central corneal area(P<0.05). At 1 mo after SPT-TPRK, the CET at 0-2 mm area had statistically significantly decreased(P<0.05). At 3 mo after SPT-TPRK, the CET at 0-2 mm had essentially reached preoperative levels. Postoperative CD values increased, with a positive correlation between CET in the 0-2 mm area and CD in the whole 0-2 mm area(r=0.256, P<0.05), and a positive correlation between CET in the 2-5 mm area and CD in the anterior 2-6 mm area(r=0.319, P<0.05).CONCLUSION: Corneal epithelial remodeling takes 3 mo in areas within 2 mm of the central cornea; areas with thinner CET have faster postoperative corneal epithelial remodeling and greater thickening in the early postoperative period; CD increases in the early postoperative period compared to the preoperative value, and in some areas, there is a positive correlation between CET and CD value.

Objective To design a medical order pre-evaluation tool for perioperative prophylactic application of antibiotics,and to improve the efficiency of antibiotics management in hospitals during the perioperative period.Methods Using the open-source software R as the platform,a web application was built with tidy verse and shiny package based on related documents and guidelines.The discharge records of Affiliated Hospital of North Sichuan Medical College from April 1,2021,to December 31,2022,were retrospectively reviewed using the constructed pre-evaluation tool and compared with previous manual evaluation results using McNemar's Chi-squared test.Results This medical order pre-evaluation tool can quickly complete perioperative antibiotics order sampling,batch pre-evaluation,result statistics,visualization,and result output,and flexibly adjust the evaluation rules according to actual needs.The pre-evaluation tool is more efficient,with a review speed of 13.46 ms per medical record.Among the 2 642 discharge medical records of manual review,there was no significant difference between systematic pre-evaluation and manual evaluation results(ratio of prophylactic use:76.85%vs.78.21%)in terms of the type of use(preventive or curative)(P= 0.078).Among the 1 857 discharge records judged to be prophylactic for both manual and systematic reviews,the difference in unreasonable detection rate(39.90%vs.30.32%)was statistically significant(P＜0.001).Among the 63 typical ludicrous medical records confirmed by the review of clinical pharmacists with senior professional titles,60 were judged and limited by the pre-evaluation tool,and the detection rate of typical unreasonable was 95.24%.Conclusions The pre-evaluation tool based on R in this study can improve the efficiency of perioperative antibacterial drug evaluation.The evaluation conclusions and statistical results are reliable and are worthy of further development and application.

[Objective]To investigate the association between screen content and the frequency of screen exposure at the age of one and a half years and hyperactive behavior in preschool,and to explore how the association is affected by the interaction between outdoor activities and screen behaviors,which could provide theoretical basis and feasible solutions for the prevention and intervention of behavioral problems in childhoood.[Methods]The survey was conducted from June 2022 to June 2023 in Huicheng District,Huizhou (China) stratified by whole cluster sampling methods. Parents and teachers of 5648 children in 61 kindergartens were sampled for questionnaire surveys. The Conners Teacher Rating Scale (TRS) was used to investigate hyperactive behavior. A self-administered questionnaire was used to investigate basic demographic information of children,screen content,frequency of screen exposure and outdoor activities at the age of one and a half years. Multivariate logistic regression was used to explore the association between video screen behavior and hyperactive behavior and its interaction with outdoor activities by controlling for covariates such as children's age,gender,and parental education.[Results]Result showed the overall prevalence of 3.2％ for hyperactive behavior,2.1％ for conduct problems,2.1％ for hyperactivity problems,1.3％ for inattention-passivity problems,and 0.9％ for hyperactivity index. After adjusting for confounding factors,multiple logistic regression analysis showed that screen exposure of "two to four times a week" at one and a half years old was associated with an increased detection rate of hyperactive behaviors in preschool children,with an estimated ORs (95％ CI) of 1.682 (1.141,2.480). Daily screen exposure was associated with increased detection rates of hyperactive behavior,conduct problems,hyperactivity issues,inattention-passivity problems,and hyperactivity index in pre-school age. The estimated ORs (95％ CI) were 2.136 (1.218,3.746),2.321 (1.185,4.546),2.300 (1.208,4.380),2.776 (1.267,6.085) and 3.640 (1.525,8.687),respectively. But the above associations were not found in children who were engaged in daily outdoor activities at the age of one and a half years (P value for interaction<0.001). No association was found between screen content and hyperactive behavior (P>0.05).[Conclusions]Frequency of screen exposure in early childhood is significantly associated with hyperactive behavior problems in preschool,and outdoor activities could weaken the correlation between high-frequency screen exposure and hyperactive behavior,suggesting that parents and schools should prioritize scientifically guiding children's video viewing behavior and outdoor activities,ensuring a well-arranged daily life,to lay a good foundation for the healthy development of children's behavior.

Objective:To explore potential predictors of refractory Mycoplasma pneumoniae pneumonia (RMPP) in early stage. Methods:The prospective multicenter study was conducted in Zhejiang, China from May 1 st, 2019 to January 31 st, 2020. A total of 1 428 patients with fever >48 hours to <120 hours were studied. Their clinical data and oral pharyngeal swab samples were collected; Mycoplasma pneumoniae DNA in pharyngeal swab specimens was detected. Patients with positive Mycoplasma pneumoniae DNA results underwent a series of tests, including chest X-ray, complete blood count, C-reactive protein, lactate dehydrogenase (LDH), and procalcitonin. According to the occurrence of RMPP, the patients were divided into two groups, RMPP group and general Mycoplasma pneumoniae pneumonia (GMPP) group. Measurement data between the 2 groups were compared using Mann-Whitney U test. Logistic regression analyses were used to examine the associations between clinical data and RMPP. Receiver operating characteristic (ROC) curves were used to analyse the power of the markers for predicting RMPP. Results:A total of 1 428 patients finished the study, with 801 boys and 627 girls, aged 4.3 (2.7, 6.3) years. Mycoplasma pneumoniae DNA was positive in 534 cases (37.4%), of whom 446 cases (83.5%) were diagnosed with Mycoplasma pneumoniae pneumonia, including 251 boys and 195 girls, aged 5.2 (3.3, 6.9) years. Macrolides-resistant variation was positive in 410 cases (91.9%). Fifty-five cases were with RMPP, 391 cases with GMPP. The peak body temperature before the first visit and LDH levels in RMPP patients were higher than that in GMPP patients (39.6 (39.1, 40.0) vs. 39.2 (38.9, 39.7) ℃, 333 (279, 392) vs. 311 (259, 359) U/L, both P<0.05). Logistic regression showed the prediction probability π=exp (-29.7+0.667×Peak body temperature (℃)+0.004×LDH (U/L))/(1+exp (-29.7+0.667×Peak body temperature (℃)+0.004 × LDH (U/L))), the cut-off value to predict RMPP was 0.12, with a consensus of probability forecast of 0.89, sensitivity of 0.89, and specificity of 0.67; and the area under ROC curve was 0.682 (95% CI 0.593-0.771, P<0.01). Conclusion:In MPP patients with fever over 48 to <120 hours, a prediction probability π of RMPP can be calculated based on the peak body temperature and LDH level before the first visit, which can facilitate early identification of RMPP.

By conducting retrospective analysis, this study aim to investigate the resistance mechanism of quinolones in non-typhoidal Salmonella (NTS). A total of 105 strains of NTS isolated from clinical specimens from the Fifth Affiliated Hospital of Southern Medical University from May 2020 to February 2021 were used as research objects. VITEK2 Compact automatic identification drug sensitivity analysis system and serological test were used to identify the strains. The sensitivity of the strains to ciprofloxacin, levofloxacin and nalidixic acid was detected by AGAR dilution method. The whole genome of 105 strains of NTS was sequenced. Abricate and other softwares were used to analyze drug-resistant genes, including plasmid-mediated quinolone resistance gene (PMQR) and Quinolone resistance determination region (QRDR). Serotypes and ST types were analyzed using SISTR and MLST, and phylogenetic trees were constructed. The results showed that the NTS isolated in this region were mainly ST34 Salmonella typhimurium (53.3%). The drug sensitivity results showed that the drug resistance rates of NTS to ciprofloxacin, levofloxacin and nalidixic acid were 30.4%, 1.9% and 22.0%, respectively, and the intermediate rates of ciprofloxacin and levofloxacin were 27.6% and 54.2%.A total of 46 (74.2%) of the 62 quinolone non-susceptible strains carried the PMQR gene, mainly qnrS1 (80.4%), followed by aac(6′)-Ib-cr(15.2%); there were 14 NTS and 8 NTS had gyrA and parC gene mutations, respectively. The gyrA was mutations at the amino acid position 87, Asp87Tyr, Asp87Asn, Asp87Gly, and Thr57Ser mutations were detected in parC. In conclusion, this study found that NTS had relatively high resistance to quinolones, carrying qnrS1 gene mainly resulted in decreased sensitivity of NTS to ciprofloxacin and levofloxacin, and gyrA:87 mutation mainly resulted in NTS resistance to Nalidixic acid; Salmonella typhimurium in clinical isolates showed clonal transmission and required further epidemiological surveillance.