Intravitreal anti-vascular endothelial growth factor(anti-VEGF)therapy has been widely used, but the variability in its therapeutic efficacy limits individualized treatment. In recent years, the application of artificial intelligence(AI)has opened up new avenues for personalized treatment response prediction, and its core branches include machine learning(ML)and deep learning(DL). This review systematically retrieved and analyzed 41 relevant studies published up to April 2025. Comprehensive analysis reveals that AI predictive models are evolving from forecasting single endpoints(such as visual acuity or central retinal thickness)to integrating multi-dimensional endpoints(encompassing anatomical, functional, and treatment demand parameters)and generating predictive imaging outputs. In terms of technical approaches, DL models(28 studies, accounting for 68.3%)dominate this field due to their robust image interpretation capabilities, while ML models(10 studies, 24.4%)retain significant value in the analysis of structured clinical data. Cross-disease comparisons indicate that research efforts are most concentrated on age-related macular degeneration(ARMD)and diabetic macular edema(DME), with shared conceptual frameworks for model construction, yet distinct anatomical and functional indicators are prioritized for each disease. Currently, the field confronts several key challenges, including insufficient prospective clinical validation, limited model interpretability(the “black box problem”), and a scarcity of high-quality multi-center datasets. Moving forward, it is imperative to advance real-world validation and develop explainable AI techniques to expedite the clinical translation of these predictive models.

Objective:To develop and validate clinical predictive models for identifying poor short-term response to recombinant human growth hormone(rhGH) treatment in children with short stature.Methods:A retrospective analysis was conducted on 118 children diagnosed with growth hormone deficiency or idiopathic short stature who were treated at the First Affiliated Hospital of Zhengzhou University and two other hospitals between January 1, 2020, and January 1, 2024. A poor response to rhGH was defined as a height increase of less than 0.2 standard deviation score(SDS) after 6 months of rhGH treatment. LASSO regression was used to identify predictive variables from baseline and follow-up data. Two logistic regression models were conducted: Model A(incorporating baseline variables only) and model B(incorporating both baseline and follow-up variables), and nomograms were created for visualization. External data and internal resampling were used for dual validation of the models, and their performance was compared.Results:A total of 118 children with short stature were included. Six baseline predictive variables(diagnosis, initial height SDS, bone age, bone age-chronological age difference, rhGH dose, and gender) and one follow-up variable(height SDS after 3 months of rhGH treatment) were identified. Area under the curve values for Model A and Model B were 0.753(95% CI 0.696-0.811) and 0.930(95% CI 0.891-0.975), respectively. Calibration curves, decision curve analysis, and other evaluation metrics demonstrated good discrimination and clinical utility for both models. Model B, incorporating the 3-month follow-up variable, showed superior predictive performance compared to Model A. Conclusions:The clinical prediction models developed in this study(Model A and Model B) are practical and reliable tools for quantitatively, conveniently, and intuitively identifying children with short stature at risk of poor response to rhGH treatment.

Objective:To investigate the effects of risperidone combined with metformin on homocysteine levels and glycolipid metabolism in patients with schizophrenia.Methods:A randomized controlled study was conducted involving 124 patients with schizophrenia who received treatment at Quzhou Third Hospital from January 2022 to June 2023. The patients were divided into a study group and a control group, with 62 patients in each group, based on a random number table. The study group was treated with risperidone combined with metformin, while the control group received risperidone treatment along with a placebo of the same dose as metformin. The treatment duration for both groups was 3 months. Clinical efficacy was evaluated, and changes in homocysteine levels and glycolipid metabolism were measured before and after treatment.Results:The difference in overall efficacy between the two groups was not statistically significant ( P = 0.697). After treatment, the homocysteine levels in the study group [(15.26 ± 2.11) μmol/L] were significantly lower than those in the control group [(19.01 ± 2.04) μmol/L, t = 9.87, P < 0.001]. The fasting blood glucose [(6.15 ± 0.57) mmol/L], 2-hour postprandial blood glucose [(12.41 ± 2.67) mmol/L], fasting insulin [(24.23 ± 3.21) μIU/mL], and total cholesterol [(6.04 ± 1.39) mmol/L] in the study group were all higher than those in the control group [(4.93 ± 0.45) mmol/L, (10.68 ± 2.46) mmol/L, (22.93 ± 3.05) μIU/mL, (0.91 ± 0.12) mmol/L, t = -13.23, -3.75, -2.31, -6.24, all P < 0.05]. The waist-to-hip ratio in the study group was (0.81 ± 0.09), which was significantly lower than that in the control group [(0.91 ± 0.12), t = 5.25, P < 0.001]. There was no significant difference in incidence of adverse reactions between the two groups ( P = 0.081). Conclusions:The combination of risperidone and metformin in the treatment of schizophrenia is beneficial for decreasing patients' homocysteine levels, preventing abnormalities in glycolipid metabolism, and reducing waist-to-hip ratios. This treatment demonstrates good efficacy and safety.

Objective:To investigate the effects of risperidone combined with metformin on homocysteine levels and glycolipid metabolism in patients with schizophrenia.Methods:A randomized controlled study was conducted involving 124 patients with schizophrenia who received treatment at Quzhou Third Hospital from January 2022 to June 2023. The patients were divided into a study group and a control group, with 62 patients in each group, based on a random number table. The study group was treated with risperidone combined with metformin, while the control group received risperidone treatment along with a placebo of the same dose as metformin. The treatment duration for both groups was 3 months. Clinical efficacy was evaluated, and changes in homocysteine levels and glycolipid metabolism were measured before and after treatment.Results:The difference in overall efficacy between the two groups was not statistically significant ( P = 0.697). After treatment, the homocysteine levels in the study group [(15.26 ± 2.11) μmol/L] were significantly lower than those in the control group [(19.01 ± 2.04) μmol/L, t = 9.87, P < 0.001]. The fasting blood glucose [(6.15 ± 0.57) mmol/L], 2-hour postprandial blood glucose [(12.41 ± 2.67) mmol/L], fasting insulin [(24.23 ± 3.21) μIU/mL], and total cholesterol [(6.04 ± 1.39) mmol/L] in the study group were all higher than those in the control group [(4.93 ± 0.45) mmol/L, (10.68 ± 2.46) mmol/L, (22.93 ± 3.05) μIU/mL, (0.91 ± 0.12) mmol/L, t = -13.23, -3.75, -2.31, -6.24, all P < 0.05]. The waist-to-hip ratio in the study group was (0.81 ± 0.09), which was significantly lower than that in the control group [(0.91 ± 0.12), t = 5.25, P < 0.001]. There was no significant difference in incidence of adverse reactions between the two groups ( P = 0.081). Conclusions:The combination of risperidone and metformin in the treatment of schizophrenia is beneficial for decreasing patients' homocysteine levels, preventing abnormalities in glycolipid metabolism, and reducing waist-to-hip ratios. This treatment demonstrates good efficacy and safety.

Objective:To develop and validate clinical predictive models for identifying poor short-term response to recombinant human growth hormone(rhGH) treatment in children with short stature.Methods:A retrospective analysis was conducted on 118 children diagnosed with growth hormone deficiency or idiopathic short stature who were treated at the First Affiliated Hospital of Zhengzhou University and two other hospitals between January 1, 2020, and January 1, 2024. A poor response to rhGH was defined as a height increase of less than 0.2 standard deviation score(SDS) after 6 months of rhGH treatment. LASSO regression was used to identify predictive variables from baseline and follow-up data. Two logistic regression models were conducted: Model A(incorporating baseline variables only) and model B(incorporating both baseline and follow-up variables), and nomograms were created for visualization. External data and internal resampling were used for dual validation of the models, and their performance was compared.Results:A total of 118 children with short stature were included. Six baseline predictive variables(diagnosis, initial height SDS, bone age, bone age-chronological age difference, rhGH dose, and gender) and one follow-up variable(height SDS after 3 months of rhGH treatment) were identified. Area under the curve values for Model A and Model B were 0.753(95% CI 0.696-0.811) and 0.930(95% CI 0.891-0.975), respectively. Calibration curves, decision curve analysis, and other evaluation metrics demonstrated good discrimination and clinical utility for both models. Model B, incorporating the 3-month follow-up variable, showed superior predictive performance compared to Model A. Conclusions:The clinical prediction models developed in this study(Model A and Model B) are practical and reliable tools for quantitatively, conveniently, and intuitively identifying children with short stature at risk of poor response to rhGH treatment.

Abstract: Objective　To study the biological function of the possible efflux pump gene BPSL1972 in Burkholderia pseudomallei (Bp), so as to improve the drug resistance mechanism of Bp. Methods BPSL1972 gene knockout strain and its complement strain were constructed by homologous recombination and electric shock transformation, respectively. BPSL1972 protein expression vector was constructed by enzyme digestion and enzyme-linked method. The BPSL1972-encoded protein was expressed using a His-tag protein expression kit and the expressed proteins were verified by Western Blot. The changes in drug sensitivity, growth, biofilm formation, and cytotoxicity were compared by microbroth dilution assay, Kirby-Bauer assay, turbidimetric assay, tube assay, crystal violet staining, and CCK-8 kit. The physicochemical properties of BPSL1972 were predicted by Protscale and Protparam online websites, the signal peptide of BPSL1972 was predicted by SignalP 5.0 online website, and the transmembrane domain and gene function of BPSL1972 were predicted by Smart online website. The 3D structure of BPSL1972-encoded protein and its homology were analyzed through Phyre 2 online website. Results The predictions suggested that the BPSL1972-encoded protein is a class of neutral-stable hydrophobic protein with a lipoprotein signal peptide and a transmembrane region, classified under the RND efflux pump system with two lipoprotein outer membrane efflux proteins, potentially mediating macrolide resistance. Its 3D modeling results show high homology to VceC and fusaric acid efflux protein, possibly mediating the efflux of ion activators. Sequencing results confirmed the successful construction of both the deletion mutant strain and complement strain. However, there were no significant differences in biological characteristics between the deletion mutant strain and the wild-type strain, and there was also no significant difference in resistance to antibiotics such as cephalosporins, quinolones, carbapenems, aminoglycosides, tetracyclines, and surfactants such as sodium deoxycholate. Conclusion BPSL1972 may be an auxiliary gene of LysR family, and this auxiliary gene is secondary, thus resulting in no functional changes after its deletion. Further research is needed to investigate whether it has resistance to other antibiotics such as fusaric acid and its specific replacement or auxiliary function.

Objective To compare the effect of glycopyrrolate or atropine in combination with neostigmine on adverse cardiovascular events(ACEs)after operation in elderly patients undergoing laparo-scopic surgery.Methods A total of 142 patients scheduled for elective laparoscopic surgery were enrolled,69 males and 73 females,aged 65-80 years,BMI 18-28 kg/m2,ASA physical status Ⅰ or Ⅱ.The pa-tients were randomly divided into two groups:the glycopyrrolate group(group G)and the atropine group(group A),71 patients in each group.After the last administration of muscle relaxants for more than 30 mi-nutes,antagonizing residual neuromuscular blockade was performed.Glycopyrrolate 4 μg/kg and neostigmine 20 μg/kg were given intravenously in group G,atropine 10 μg/kg and neostigmine 20 μg/kg were given intravenously in group A.The incidence of ACEs and severe ACEs during operation and 72 hours after operation were recorded.Recovery situation in PACU such as NRS scores at rest and coughing,Rich-mond agitation-sedation scale(RASS)score,and modified Aldrete score 15 and 30 minutes after extubation were recorded.Emergence agitation,dry mouth,nausea,vomiting,and delirium 24 hours after operation were recorded.Results Compared with group A,the total incidence of ACEs,tachycardia,and myocardial ischemia after operation were significantly decreased in group G(P＜0.05),the incidence of dry mouth 24 hours postoperatively was significantly increased in group G(P＜0.05).There was no severe ACEs oc-curred in the two groups 72 hours after operation.Conclusion Compared with atropine,glycopyrrolate combined with neostigmine in elderly patients undergoing laparoscopic surgery can reduce the incidence of cardiac tachycardia,myocardial ischemia,and total ACEs after operation,and there was no severe ACEs occurred.However,it can increase the incidence of dry mouth 24 hours after operation.

Objective:To explore the incidence of thirst in patients undergoing pancreaticoduodenectomy and analyze its influencing factors.Methods:From March to October 2022, 155 patients with pancreaticoduodenectomy admitted to the First Medical Center of the Chinese People's Liberation Army General Hospital were selected as the study subject by convenience sampling. The patients were surveyed using the self-made general information questionnaire, Thirst Numerical Rate Scale, oral mucosal misture score, Self-Rating Anxiety Scale, and Self-Rating Depression Scale to understand the incidence of thirst and analyze influencing factors.Results:A total of 153 valid questionnaires were collected, with a valid response rate of 98.71%. Among them, 124 patients (81.05%) experienced postoperative thirst. Univariate analysis showed that there were statistical differences in the incidence of thirst among patients with different ages, fasting time, smoking history, drinking history, hypertension history, diabetes history, anxiety and depression ( P<0.05). Logistic regression analysis showed that age, hypertension history, diabetes history, anxiety and depression were the main influencing factors for thirst in patients undergoing pancreaticoduodenectomy ( P<0.05) . Conclusions:Thirst is common in patients after pancreaticoduodenectomy. Age, hypertension history, diabetes history, anxiety and depression are the influencing factors of thirst in patients after pancreaticoduodenectomy. Medical and nursing staff should provide targeted interventions for thirst prevention and treatment based on relevant factors.

Abstract： Objective　To analyze the characteristics and corresponding drug resistance of pathogenic bacterial spectrum in eight major infection sites of hospitalized patients, and to provide epidemiological data for the rational selection of antibiotics in clinical practice. Methods　A total of 396 bacterial strains isolated from clinical specimens of hospitalized patients in member institutions of the Hainan Provincial Bacterial Resistance Monitoring Network from September 1, 2020, to September 30, 2022, were included in this study. Data were screened and filtered from the database of MH120 Microbial Identification and Drug Sensitivity Analysis System based on the technical scheme of the National Bacterial Drug Resistance Surveillance Network and Science and Technology Basic Resources Investigation Project research plan in 2020. The testing data were integrated, summarized, and analyzed using EXCEL and WHONET 5.6 software, and statistical analysis was conducted using SPSS 26.0 software. Results　Among of 396 strains of bacteria, 78 (19.7%) were isolated from respiratory tract specimens, 74 (18.7%) from urinary tract specimens, 72 (18.2%) from blood specimens, 54 (13.6%) from abdominal cavity specimens, 48 (12.1%) from skin and soft tissue specimens 48 strains (12.1%), 30 (7.6%) from reproductive tract specimens, 22 (5.6%) from central nervous system specimens, 18 (4.5%) from digestive tract specimens. Gram-negative bacteria accounted for 69.4% of the isolates, while gram-positive bacteria accounted for 30.6%. The top five gram-negative bacteria isolated were Klebsiella pneumoniae (14.9%), Escherichia coli (14.4%), Pseudomonas aeruginosa (10.4%), Acinetobacter baumannii (5.3%), and Salmonella species (4.5%). The top five gram-positive bacteria were Staphylococcus aureus (11.1%), Streptococcus agalactis (7.8%), Enterococcus faecalis (3.0%), Enterococcus faecium (2.8%), and Streptococcus suis (1.8%). Respiratory failure and bloodstream infection were independent influencing factors of treatment response (P<0.01). The resistance rate of Escherichia coli to ampicillin was 81.4%, and the resistance rate of Staphylococcus aureus to gentamicin and levofloxacin were both below 7%. Conclusions　The pathogen spectra vary with different infection sites of patients, and rational selection of antibiotics based on drug susceptibility testing is crucial to shorten the treatment time of patients and avoid the unnecessary emergence of drug-resistant strains caused by drug abuse.

Objective:To investigate the application value of contrast-enhanced ultrasound (CEUS) before and after microwave ablation of thyroid nodules.Methods:Fifty-six patients (79 thyroid nodules) who received microwave ablation of thyroid nodules in Huaian Medical District, General Hospital of Eastern Theater Command from March 2016 to October 2019 were included in this study. CEUS was performed before microwave ablation to accurately assess the size, number and blood supply of thyroid nodules as well as the position of the feeding vessels. CEUS was performed immediately after microwave ablation to determine whether the lesion area was thoroughly ablated and to measure the volume of thyroid nodules. At 1, 3, 6 and 12 months after surgery, the level of thyroid hormone was measured and the absorption of thyroid nodules was evaluated.Results:Preoperative CEUS showed that among the 79 thyroid nodules, 42 were solid nodules that had different degrees of enhancement, including 33 annular homogeneously highly enhanced nodules and 9 heterogeneously highly enhanced nodules; 24 were cystic mixed solid nodules that had solid components, including 16 homogeneously highly enhanced nodules and 8 nodules with only local high enhancement in the solid component; 13 were cystic nodules, including 9 nodules with septa and 3 nodules with contrast medium on the diaphragm. Contrast medium was still visible around three nodules immediately after microwave ablation. Ablation continued in three nodules until there was no contrast medium. The incidence of complications during and after treatment was 0%. The average volume of the thyroid nodules before treatment was (7.52 ± 6.74) cm3. At 1, 3, 6 and 12 months after surgery, the average volume of the thyroid nodules was (6.06 ± 5.19) cm3, (3.06 ± 2.85) cm3, (1.32 ± 1.23) cm3 and (0.59 ± 0.52) cm 3, respectively. There was significant difference in volume of thyroid nodules between before and after microwave ablation ( F = 96.32, P < 0.001). Conclusion:Preoperative CEUS can determine the distribution of the blood supply of thyroid nodules and the course of the feeding vessels, identify the needle-entering position for microwave ablation and the primary ablation area, improve the accuracy of treatment, and reduce the occurrence of complications such as bleeding. Postoperative CEUS can determine whether lesion area is thoroughly ablated, reduce residual lesions and excessive ablation.