OBJECTIVE To construct a predictive model for evaluating the efficacy of a triple antiemetic regimen （neurokinin- 1 receptor antagonist+5-hydroxytryptamine 3 receptor antagonist+dexamethasone） for preventing nausea and vomiting induced by highly emetogenic chemotherapy （HEC） based on interpretable deep learning algorithms. METHODS Clinical data of cancer patients who received HEC and were treated with the standard triple antiemetic regimen in the oncology department of Tianjin Medical University General Hospital from January 2018 to December 2022 were collected retrospectively. Demographic， clinical and metabolism-related variables were integrated. After data pre-processing， two deep learning algorithms （deep random forest and dense neural network） and four machine learning algorithms （support vector machine， categorical boosting， random forest and decision tree） were used to build predictive models. Subsequently， model performance evaluation and model interpretability analysis were conducted. RESULTS Among the six candidate models， the deep random forest model demonstrated the best predictive performance on the test set， with an area under the receiver operating characteristic curve of 0.850， an accuracy of 0.911， a precision of 0.805， a recall of 0.783， an F1 score of 0.793， and a Brier score of 0.075. Interpretability analysis revealed that creatinine clearance rate （Ccr） was the key predictive factor， and low Ccr levels， female gender， younger age， highly emetogenic drugs （particularly cisplatin-containing chemotherapy regimens）， and anticipatory nausea and vomiting were positively correlated with the risk of HEC-related nausea and vomiting. CONCLUSIONS The deep random forest model exhibits the best performance in predicting the efficacy of triple antiemetic regimen for preventing HEC-related nausea and vomiting. The key predictors in this model primarily include Ccr，anticipatory nausea and vomiting， gender， age， and highly emetogenic drugs.

At present, the treatment of sepsis depends largely on non-specific methods, highlighting an urgent need for novel therapeutic strategies. Stem cells have garnered significant attention in the treatment of various diseases due to their unique biological properties. Stem cells enhance sepsis survival through mechanisms such as reducing bacterial burden, modulating inflammation, and ameliorating organ dysfunction. Recent studies have shown that stem cells can increase the survival rate of sepsis patients through multiple pathways such as reducing the bacterial load of the host, regulating inflammatory homeostasis, and improving multi-organ dysfunction. Their derivatives, exosomes, can also alleviate the imbalanced immune response in sepsis patients. Recent advances in stem cell-based therapies for sepsis were summarized in this paper.

Background Meteorological factors are among the key extrinsic triggers for the onset and exacerbation of cardiovascular and cerebrovascular diseases (CVD). Against the backdrop of sustained global warming, elucidating the impact of ambient temperature and atmospheric pressure on CVD, especially on pre-hospital CVD emergent events, has become imperative for evidence-based prevention and emergency preparedness. Objective To quantify the temporal trends of daily mean temperature and atmospheric pressure and their associations with pre-hospital CVD emergent events in Yantai, and to explore effect modification by demographic subgroups and geographic areas, thereby providing an empirical basis for the rational allocation of emergency medical resources. Methods Pre-hospital CVD emergency data from January 1, 2016 to December 31, 2022 were selected from the Yantai 120 Emergency Medical Command System. Synchronous meteorological factors and environmental pollutant data were obtained from the websites of the National Oceanic and Atmospheric Administration and the National Centers for Environmental Information of the United States. Time-series analysis combined with distributed lag non-linear model was used to analyze the association between daily temperature, atmospheric pressure, and pre-hospital CVD emergencies. Average annual percentage changes (AAPC) were calculated using Joinpoint (version 5.2.0.0) to reflect temporal trends. Spearman correlation analysis was employed to screen variables with low collinearity for inclusion in the multi-pollutant adjusted models. Results From 2016 to 2022, a total of 268275 pre-hospital CVD emergency cases were recorded in Yantai City, showing an increasing annual trend (AAPC=7.16%, P=0.002). Emergency volume, daily temperature, atmospheric pressure, and PM₁₀ displayed marked seasonal patterns, with summer bottoms and winter peaks for emergencies, daily temperature higher in summer and lower in winter, and the inverse for atmospheric pressure. Correlation analysis showed that both daily temperature and atmospheric pressure were positively correlated with O₃, with correlation coefficients of 0.745 and 0.723, respectively; negatively correlated with PM_2.5, with correlation coefficients of −0.246 and −0.235, respectively; and negatively correlated with PM₁₀, with correlation coefficients of −0.241 and −0.229, respectively (P < 0.05). A U-shaped correlation was revealed between daily temperature and pre-hospital CVD emergency risk. Using 24 °C as the reference, risk was elevated at both low and high daily temperatures, with the strongest cold effect at –11 ℃ (RR=1.53, 95%CI: 1.35, 1.73) and strongest heat effect at 30 ℃ (RR=1.06, 95%CI: 1.01, 1.11); cold effects were prolonged (lag0–14d) compared to heat effects. Atmospheric pressure exhibited a positive association with emergencies; using a median of 1012.2 hPa as the reference, the maximum risk occurred at 1037 hPa (RR=1.25, 95%CI: 1.03, 1.52), with a 0–5 d lag. In the subgroup analysis, the impact of low temperature was more pronounced among females (RR_{−9 ℃}=1.66), people over 60 years (RR_{−9 ℃}=1.91), and rural residents (RR_{−9 ℃}=1.76). High atmospheric pressure significantly increased the risk of pre-hospital CVD emergency among females (RR_{1035 hPa}=1.54), people under 18 years (RR_{1035 hPa}=3.62), and rural residents (RR_{1035 hPa}=1.46). The sensitivity analysis results indicated that, after excluding the impact of pandemic and the degrees of freedom for time variations, the effects of daily average temperature and atmospheric pressure on pre-hospital CVD emergency volume were consistent with the primary analysis findings. Conclusion Both non-optimal temperature and atmospheric pressure significantly increase the pre-hospital CVD emergency volume; therefore, emergency medical services should develop targeted response protocols, optimize resource allocation, and intensify public health education to mitigate the pre-hospital burden attributable to temperature and pressure fluctuations.

By analyzing the current status of randomized controlled trials （RCTs） on functional dyspepsia （FD） and irritable bowel syndrome （IBS） comorbidity， we identified several critical issues which include insufficient repor-ting of FD and IBS subtypes， inadequate risk assessment of drug combination， lack of composite， objective， and long-term outcomes， and weak evidence support for clinical practice guidelines. It is suggested that future clinical research on FD-IBS comorbidity should further strengthen the application of real-world studies. The use of composite outcomes and long-term follow-up is recommended to improve the quality of evidence， while greater attention should be paid to patients' preferences and self-management to enhance the applicability of findings. Based on the existing issues in clinical studies of traditional Chinese medicine （TCM） for FD-IBS comorbidity， we propose to consolidate the foundation of TCM-specific efficacy evaluation to better reflect the advantages of syndrome differentiation and treatment， optimize real-world study designs to better support clinical decision- making， and introduce intelligent objective evaluation technologies to improve the objectivity and accuracy of TCM clinical efficacy assessment.

By analyzing the current status of randomized controlled trials （RCTs） on functional dyspepsia （FD） and irritable bowel syndrome （IBS） comorbidity， we identified several critical issues which include insufficient repor-ting of FD and IBS subtypes， inadequate risk assessment of drug combination， lack of composite， objective， and long-term outcomes， and weak evidence support for clinical practice guidelines. It is suggested that future clinical research on FD-IBS comorbidity should further strengthen the application of real-world studies. The use of composite outcomes and long-term follow-up is recommended to improve the quality of evidence， while greater attention should be paid to patients' preferences and self-management to enhance the applicability of findings. Based on the existing issues in clinical studies of traditional Chinese medicine （TCM） for FD-IBS comorbidity， we propose to consolidate the foundation of TCM-specific efficacy evaluation to better reflect the advantages of syndrome differentiation and treatment， optimize real-world study designs to better support clinical decision- making， and introduce intelligent objective evaluation technologies to improve the objectivity and accuracy of TCM clinical efficacy assessment.

AIM: To investigate the distribution characteristics of pathogens in patients with post-traumatic infectious endophthalmitis(PTIE)and their relationship with serum levels of macrophage inflammatory protein 1α(MIP-1α), heat shock protein 70(HSP70), and soluble triggering receptor expressed on myeloid cells 1(sTREM-1).METHODS:A total of 157 patients with PTIE from the Handan City Eye Hospital(The Third Hospital of Handan)from May 2023 to May 2025 were selected as the study group. They were divided into a good prognosis group and a poor prognosis group based on their uncorrected visual acuity at discharge. Meanwhile, 157 patients with ocular trauma but without endophthalmitis during the same period were selected as control group 1, and 157 healthy volunteers who underwent physical examinations during the same period were selected as control group 2. Aqueous humor and vitreous fluid samples were collected from the study group to detect the distribution of pathogens. The levels of serum MIP-1α, HSP70, and sTREM-1 were measured using the enzyme-linked immunosorbent assay. Multivariate Logistic regression analysis was performed to identify risk factors for poor prognosis. The predictive value of serum MIP-1α, HSP70, and sTREM-1 levels for poor prognosis was evaluated using receiver operating characteristic(ROC)and decision curve analysis(DCA).RESULTS: The general data of the participants in the three groups was comparable. A total of 173 pathogens were detected in the 157 patients with PTIE, with Gram-positive bacteria being the predominant type. The levels of serum MIP-1α and sTREM-1 in the study group were higher than those in control groups 1 and 2, while the level of HSP70 was lower than those in control groups 1 and 2(all P<0.05). There were no significant differences in the levels of serum MIP-1α, HSP70, and sTREM-1 between control groups 1 and 2(all P>0.05). In the poor prognosis group, the time of wound suture was ≥24 h, the wound location was in zones II/III, the type of trauma was rupture, the proportion of rupture injuries, and the levels of serum C-reactive protein, MIP-1α, and sTREM-1 were higher than those in the good prognosis group, while the level of HSP70 was decreased(all P<0.001). Multivariate Logistic regression analysis showed that the time of wound suture, wound location, type of trauma, C-reactive protein, MIP-1α, HSP70, and sTREM-1 were risk factors for poor visual prognosis in patients with PTIE(all P<0.05). The ROC curve results showed that the combined prediction of serum MIP-1α, HSP70, and sTREM-1 for poor visual prognosis in PTIE patients had an AUC value of 0.965, which was significantly higher than that of individual predictions(ZMIP-1α, ZHSP70, ZsTREM-1=3.628, 4.705, 3.930, all P<0.05). Additionally, the DCA curve showed that the combined prediction had a higher net benefit rate than individual predictions in the high-risk threshold range of 0.03-0.97.CONCLUSION:Gram-positive bacteria are the predominant type of pathogenic bacteria in patients with PTIE, with elevated levels of serum MIP-1α and sTREM-1 and decreased levels of HSP70. The combined detection of these three factors has a high predictive efficacy for visual prognosis in patients.

ObjectiveTo evaluate the impact of stroke on diaphragmatic function by ultrasound. MethodsFrom April, 2024 to June, 2025, 40 stroke patients (stroke group) were recruited in Xuanwu Hospital Capital Medical University. Additionally, 40 sex- and age-matched healthy subjects (control group) were recruited without directional selection. Diaphragmatic excursion and diaphragmatic thickness during quiet breathing and deep breathing were assessed using ultrasound in both groups, and the diaphragmatic thickening fraction was calculated. The stroke group was further divided into two subgroups according to disease course: ≤ 30 days group and > 30 days group, for subgroup analysis. ResultsDuring both quiet and deep breathing, diaphragmatic excursion on the affected side in the stroke group was smaller than that on the unaffected side and in the control group (P < 0.05). During quiet breathing, diaphragmatic thickening fraction on the affected side in the stroke group was smaller than that on the unaffected side and in the control group (P < 0.05), while the diaphragmatic thickening fraction on the unaffected side was significantly smaller than that in the control group (P < 0.01). During deep breathing, the bilateral diaphragmatic thickening fraction in the stroke group was significantly smaller than that in the control group (P < 0.001), and the bilateral end-inspiratory diaphragmatic thickness was significantly smaller than that in the control group (P < 0.001). During quiet breathing, diaphragmatic excursion in both the ≤ 30 d group and the > 30 d group was smaller than that in the control group (P < 0.05); during deep breathing, diaphragmatic excursion in the ≤ 30 d group was smaller than that in the control group (P < 0.05). In both quiet breathing and deep breathing, the diaphragmatic thickening fraction in the ≤ 30 d group and the >30 d group were lower than those in the control group (P < 0.05). During deep breathing, end-inspiratory diaphragmatic thickness in both the ≤ 30 d group and the > 30 d group was significantly smaller than that in the control group (P < 0.01). ConclusionUltrasound can effectively identify the impact of stroke on patients' diaphragmatic function. In stroke patients, both diaphragmatic excursion and contractile performance are reduced on the affected side, while contractile performance is also reduced on the unaffected side.

ObjectiveTo improve the therapeutic effect of Buyang Huanwutang（BYHW） on diabetic kidney disease （DKD） and explore new methods for developing new Chinese medicine decoctions，we utilized 5-hydroxymethylcytosine （5hmC）-Seal sequencing technology and network pharmacology to modify BYHW. MethodsWe selected 14 diabetes mellitus （DM） patients and 15 DKD patients hospitalized in the Department of Endocrinology of Peking University Third Hospital in 2021. Circulating free DNA （cfDNA） in the patients’ plasma was sequenced. After data processing and screening， we performed temporal clustering analysis to select a DKD 5hmC gene set， which was then cross-validated with a DKD database gene set to obtain the DKD gene set. We retrieved target genes of the seven herbal components of BYHW from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform （TCMSP） and the Encyclopedia of Traditional Chinese Medicine （ETCM）， and performed cross-analysis with the DKD gene set to identify common genes shared by the disease and the Chinese medicines. A protein-protein interaction （PPI） network was constructed for the common genes to screen out the key genes. Chinese medicines targeting these key genes were searched against ETCM to identify removable Chinese medicines. Kyoto Encyclopedia of Genes and Genomes （KEGG） enrichment analysis was performed on non-common DKD genes， and key genes in DKD-related pathways were selected based on machine learning. The GSE30529 dataset was used to verify the expression trends of 5hmC-modified genes and the feasibility of target genes as drug targets. TCMBank was used to search for target genes and obtain compounds targeting these genes and the corresponding Chinese medicines to construct a "key target-compound-Chinese medicine" network. Molecular docking was employed to verify the binding affinity of compounds with key targets. TCMSP and ETCM were used to search and count the candidate Chinese medicines targeting DKD-related genes， and a new decoction was formed by adding the selected Chinese medicines. A mouse model of DKD was established to examine the efficacy of the new decoction based on the mouse body mass， random blood glucose， urinary microalbumin （mALB）， serum creatinine （Scr）， and blood urea nitrogen （BUN） and by hematoxylin-eosin staining， periodic acid-Schiff staining， Masson staining， immunofluorescence assay， and Real-time PCR. ResultsThe cross-analysis results showed that the DKD gene set included 507 genes， of which 30 were target genes of BYHW. The PPI analysis indicated that the top 15% target genes regarding the degree were interleukin-6 （IL-6）， Toll-like receptor 4 （TLR4）， lactotransferrin （LTF）， lipoprotein lipase （LPL）， and sterol regulatory element-binding transcription factor 1 （SREBF1）. Persicae Semen and Pheretima in BYHW were unrelated to key genes and removed. Machine learning identified 10 potential target genes， among which TBC1 domain family member 5 （TBC1D5）， RAD51 paralog B （RAD51B）， and proteasome 20S subunit alpha 6 （PSMA6） had expression trends consistent with the GSE30529 dataset and could serve as drug targets. The "key target-compound-Chinese medicine" network and molecular docking results indicated that the compounds with good binding affinity to target proteins were arginine， glycine， myristicin， serine， and tyrosine， corresponding to 121 Chinese medicines. The top 10 Chinese medicines targeting DKD-related genes were Lycii Fructus， Ginseng Radix et Rhizoma， Dioscoreae Rhizoma， Rehmanniae Radix Praeparata， Isatidis Radix， Glehniae Radix， Ophiopogonis Radix， Allii Sativi Bulbus， Isatidis Folium， and Bolbostemmatis Rhizoma. Based on traditional Chinese medicine theory， the new decoction was obtained after removal of Persicae Semen and Pheretima and addition of Rehmanniae Radix Praeparata and Dioscoreae Rhizoma. Animal experiment results indicated that the modified BYHW improved the kidney function and inhibited renal fibrosis in DKD mice， with better effects than the original decoction. ConclusionThe BYHW modified based on 5hmC-Seal sequencing demonstrates better performance in inhibiting fibrosis and ameliorating DKD than the original decoction. This elucidates the biomedical theory behind the epigenetic modification of traditional Chinese medicine prescriptions， potentially offering new perspectives for the exploration of these prescriptions

ObjectiveTo analyze the adverse reactions associated with the clinical use of Banxia （Rhizoma Pinelliae）- Wutou （Aconitum） in the same formula， with the aim of providing a reference for the safety of their clinical application. MethodsLiterature on the clinical application of antagonistic herbs "Banxia-Wutou" used in the same formula， published from January 1st， 2014， to June 30th， 2023， was retrieved from databases including CNKI， VIP， Wanfang， SinoMed， PubMed， Cochrane Library， and Embase. A database was established， and information related to adverse reactions was extracted， including descriptions， classifications， specific manifestations， management and outcomes， patients' primary diseases （western medicine diseases and traditional Chinese medicine diagnoses and syndromes）， and medication information （dosage， ratio， administration routes， and dosage forms）. ResultsA total of 79 researches simultaneously used antagonistic herbs Banxia-Wutou in the same formula and reported associated advers reactions. Gastrointestinal adverse reactions were the most common， with 8 studies reporting management of adverse reactions and 3 studies reporting improvement with no intervention. Among the 11 researches， the adverse reaction relieved to extant， while other 69 researches didn't report the managment of adverse reaction and its prognosis. For the primary disease in western medicine system， chronic bronchitis and chronic obstructive pulmonary disease （COPD） were most common， while gastric pain was the most common symptom in traditional Chinese medicine with spleen and kidney deficiency and spleen stomach cold deficiency being the most frequent syndromes. The most common Banxia dosage was 10 g， while for the Wutou， Fuzi （Radix Aconiti Lateralis Praeparata） was predominant with the highest dose at 15 g. The most frequent herbal combination was Banxia-fuzi， with a 1∶1 ratio. The main administration route was oral， and the primary dosage form was decoction. ConclusionGastrointestinal adverse reactions are the most common in the clinical use of Banxia-Wutou antagonistic herb combinations. Research on the safety of "Banxia-Wutou" combinations should focus on respiratory system diseases and spleen-stomach related conditions.

ObjectiveTo improve the therapeutic effect of Buyang Huanwutang（BYHW） on diabetic kidney disease （DKD） and explore new methods for developing new Chinese medicine decoctions，we utilized 5-hydroxymethylcytosine （5hmC）-Seal sequencing technology and network pharmacology to modify BYHW. MethodsWe selected 14 diabetes mellitus （DM） patients and 15 DKD patients hospitalized in the Department of Endocrinology of Peking University Third Hospital in 2021. Circulating free DNA （cfDNA） in the patients’ plasma was sequenced. After data processing and screening， we performed temporal clustering analysis to select a DKD 5hmC gene set， which was then cross-validated with a DKD database gene set to obtain the DKD gene set. We retrieved target genes of the seven herbal components of BYHW from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform （TCMSP） and the Encyclopedia of Traditional Chinese Medicine （ETCM）， and performed cross-analysis with the DKD gene set to identify common genes shared by the disease and the Chinese medicines. A protein-protein interaction （PPI） network was constructed for the common genes to screen out the key genes. Chinese medicines targeting these key genes were searched against ETCM to identify removable Chinese medicines. Kyoto Encyclopedia of Genes and Genomes （KEGG） enrichment analysis was performed on non-common DKD genes， and key genes in DKD-related pathways were selected based on machine learning. The GSE30529 dataset was used to verify the expression trends of 5hmC-modified genes and the feasibility of target genes as drug targets. TCMBank was used to search for target genes and obtain compounds targeting these genes and the corresponding Chinese medicines to construct a "key target-compound-Chinese medicine" network. Molecular docking was employed to verify the binding affinity of compounds with key targets. TCMSP and ETCM were used to search and count the candidate Chinese medicines targeting DKD-related genes， and a new decoction was formed by adding the selected Chinese medicines. A mouse model of DKD was established to examine the efficacy of the new decoction based on the mouse body mass， random blood glucose， urinary microalbumin （mALB）， serum creatinine （Scr）， and blood urea nitrogen （BUN） and by hematoxylin-eosin staining， periodic acid-Schiff staining， Masson staining， immunofluorescence assay， and Real-time PCR. ResultsThe cross-analysis results showed that the DKD gene set included 507 genes， of which 30 were target genes of BYHW. The PPI analysis indicated that the top 15% target genes regarding the degree were interleukin-6 （IL-6）， Toll-like receptor 4 （TLR4）， lactotransferrin （LTF）， lipoprotein lipase （LPL）， and sterol regulatory element-binding transcription factor 1 （SREBF1）. Persicae Semen and Pheretima in BYHW were unrelated to key genes and removed. Machine learning identified 10 potential target genes， among which TBC1 domain family member 5 （TBC1D5）， RAD51 paralog B （RAD51B）， and proteasome 20S subunit alpha 6 （PSMA6） had expression trends consistent with the GSE30529 dataset and could serve as drug targets. The "key target-compound-Chinese medicine" network and molecular docking results indicated that the compounds with good binding affinity to target proteins were arginine， glycine， myristicin， serine， and tyrosine， corresponding to 121 Chinese medicines. The top 10 Chinese medicines targeting DKD-related genes were Lycii Fructus， Ginseng Radix et Rhizoma， Dioscoreae Rhizoma， Rehmanniae Radix Praeparata， Isatidis Radix， Glehniae Radix， Ophiopogonis Radix， Allii Sativi Bulbus， Isatidis Folium， and Bolbostemmatis Rhizoma. Based on traditional Chinese medicine theory， the new decoction was obtained after removal of Persicae Semen and Pheretima and addition of Rehmanniae Radix Praeparata and Dioscoreae Rhizoma. Animal experiment results indicated that the modified BYHW improved the kidney function and inhibited renal fibrosis in DKD mice， with better effects than the original decoction. ConclusionThe BYHW modified based on 5hmC-Seal sequencing demonstrates better performance in inhibiting fibrosis and ameliorating DKD than the original decoction. This elucidates the biomedical theory behind the epigenetic modification of traditional Chinese medicine prescriptions， potentially offering new perspectives for the exploration of these prescriptions