AIM: To evaluate the application value of artificial intelligence-assisted systems in diagnosing diabetic retinopathy(DR)by Meta-analysis.METHODS: PubMed, Web of Science, Embase, Cochrane Library, CBM, CNKI, WanFang Data and VIP database were searched to collect relevant literature on the diagnostic value of artificial intelligence-assisted systems for DR from January 2019 to September 2024. The QUADAS-2 tool was used to evaluate the quality of the included studies, and Meta-analysis was performed using Stata 17.0 and Meta Disc 1.4 software.RESULTS: A total of 23 studies were included. The results of Meta-analysis showed that the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio were 0.92(95% CI: 0.89-0.94), 0.94(95% CI: 0.91-0.96), 15.6(95% CI: 10.6-22.9), 0.09(95% CI: 0.07, 0.12), 174(95% CI: 112-271), respectively, and the area under the ROC curve(AUC)was 0.97(95% CI: 0.96-0.98). Meta-regression and subgroup analyses indicated that the heterogeneity of the studies originated from study type, patient type, patient source, and AI algorithm type. Deeks' funnel plot test suggested no significant publication bias(P=0.15), indicating that the results were robust.CONCLUSION: The artificial intelligence-assisted system demonstrates high diagnostic value for DR, and can be widely implemented in the early screening and diagnosis of DR.

Clinical chemotherapy for prostate cancer is still compromised by high treatment thresholds and severe off-target toxicity of drugs. Given the limited progress in improving therapeutic outcomes and reducing toxicity with the existing toolbox, efforts to broaden the chemotherapeutic window are highly desired. Here, we discover that gossypol (GSP, a natural compound) dramatically enhances the chemosensitivity of cabazitaxel (CTX), even at previously ineffective concentrations. Based on this interesting finding, we exploit a carrier-free chemotherapeutic nano-booster for prostate cancer treatment, which is molecularly co-assembled by GSP and cabazitaxel (CTX). GSP not only readily forms nanoassembly with CTX, but also functions as a chemotherapeutic enhancer that unlocks an ultra-low-dose chemotherapeutic window. Not only that, precise dual-drug nanoassembly confers CTX a significantly larger maximum tolerable dose. As expected, the nano-booster exerts striking therapeutic benefits in mouse prostate tumor xenograft models. This study advances chemotherapeutic window expansion and self-sensitized chemotherapy toward clinical applicability.

Triple-negative breast cancer is therapeutically challenging due to the low expression of tumor markers and 'cold' tumor immunosuppressive microenvironment. Here, we present a dual-targeting peptide-drug conjugate (PDC) for tumor inhibition. Our PDC efficiently and selectively delivers cytotoxic Monomethyl Auristatin E (MMAE) into tumor cells via C-X-C chemokine receptor type 4 (CXCR4) and folate receptor 1 (FOLR1) for synergistic inhibition of growth and metastasis. Our results show that the dual-targeting PDC has potent antitumor activity in cultured human cells and several murine transplanted tumor models without apparent toxicity. The combination of dual-targeting PDC and radiotherapy modulates the tumor immunosuppressive microenvironment by increasing CD8⁺ T cell infiltration and attenuating the proportion of myeloid-derived suppressor and regulatory T cells. Therefore, our dual-targeting PDC represents a promising new strategy for cancer therapy that rebalances the immune system and promotes tumor regression.

Depression, a prevalent mental disorder with significant socioeconomic burdens, underscores the urgent need for safe and effective non-pharmacological interventions. Recent advances in microbiome research have revealed the pivotal role of gut microbiota dysbiosis in the pathogenesis of depression. Concurrently, exercise, as a cost-effective and accessible intervention, has demonstrated remarkable efficacy in alleviating depressive symptoms. This comprehensive review synthesizes current evidence on the interplay among exercise, gut microbiota modulation, and depression, elucidating the mechanistic pathways through which exercise ameliorates depressive symptoms via the microbiota-gut-brain (MGB) axis. Depression is characterized by gut microbiota alterations, including reduced alpha and beta diversity, depletion of beneficial taxa (e.g., Bifidobacterium, Lactobacillus, and Coprococcus), and overgrowth of pro-inflammatory and pathogenic bacteria (e.g., Morganella, Klebsiella, and Enterobacteriaceae). Metagenomic analyses reveal disrupted metabolic functions in depressive patients, such as diminished synthesis of short-chain fatty acids (SCFAs), impaired tryptophan metabolism, and dysregulated bile acid conversion. For instance, Bifidobacterium longum deficiency correlates with reduced synthesis of neuroactive metabolites like homovanillic acid, while decreased Coprococcus abundance limits butyrate production, exacerbating neuroinflammation. Furthermore, elevated levels of indole derivatives from Clostridium species inhibit serotonin (5-HT) synthesis, contributing to depressive phenotypes. These dysbiotic profiles disrupt the MGB axis, triggering systemic inflammation, neurotransmitter imbalances, and hypothalamic-pituitary-adrenal (HPA) axis hyperactivity. Exercise exerts profound effects on gut microbiota composition, diversity, and metabolic activity. Longitudinal studies demonstrate that sustained aerobic exercise increases alpha diversity, enriches SCFA-producing genera (e.g., Faecalibacterium prausnitzii, Roseburia, and Akkermansia), and suppresses pathobionts (e.g., Desulfovibrio and Streptococcus). For example, a meta-analysis of 25 trials involving 1 044 participants confirmed that exercise enhances microbial richness and restores the Firmicutes/Bacteroidetes ratio, a biomarker of metabolic health. Notably, endurance training promotes Veillonella proliferation, which converts lactate into propionate, enhancing energy metabolism and delaying fatigue. Exercise also strengthens intestinal barrier integrity by upregulating tight junction proteins (e.g., ZO-1, occludin), thereby reducing lipopolysaccharide (LPS) translocation and systemic inflammation. However, excessive exercise may paradoxically diminish microbial diversity and exacerbate intestinal permeability, highlighting the importance of moderate intensity and duration. Exercise ameliorates depressive symptoms through multifaceted interactions with the gut microbiota, primarily via 4 interconnected pathways. First, exercise mitigates neuroinflammation by elevating anti-inflammatory SCFAs such as butyrate, which suppresses NF-κB signaling to attenuate microglial activation and oxidative stress in the hippocampus. Animal studies demonstrate that voluntary wheel running reduces hippocampal TNF‑α and IL-17 levels in stress-induced depression models, while fecal microbiota transplantation (FMT) from exercised mice reverses depressive behaviors by modulating the TLR4/NF‑κB pathway. Second, exercise regulates neurotransmitter dynamics by enriching GABA-producing Lactobacillus and Bifidobacterium, thereby counteracting neuronal hyperexcitability. Aerobic exercise also enhances the abundance of Lactobacillus plantarum and Streptococcus thermophilus, which facilitate 5-HT and dopamine synthesis. Clinical trials reveal that 12 weeks of moderate exercise increases fecal Coprococcus and Blautia abundance, correlating with improved 5-HT bioavailability and reduced depression scores. Third, exercise normalizes HPA axis hyperactivity by reducing cortisol levels and restoring glucocorticoid receptor sensitivity. In rodent models, chronic stress-induced corticosterone elevation is reversed by probiotic supplementation (e.g., Lactobacillus), which enhances endocannabinoid signaling and hippocampal neurogenesis. Furthermore, exercise upregulates brain-derived neurotrophic factor (BDNF) via microbial metabolites like butyrate, promoting histone acetylation and synaptic plasticity. FMT experiments confirm that exercise-induced microbiota elevates prefrontal BDNF expression, reversing stress-induced neuronal atrophy. Fourth, exercise reshapes microbial metabolic crosstalk, diverting tryptophan metabolism toward 5-HT synthesis instead of neurotoxic kynurenine derivatives. Butyrate inhibits indoleamine 2,3-dioxygenase (IDO), a key enzyme in the kynurenine pathway linked to depression. Concurrently, exercise-induced Akkermansia enrichment enhances mucin production, fortifies the gut barrier, and reduces LPS-driven neuroinflammation. Collectively, these mechanisms underscore exercise as a potent modulator of the microbiota-gut-brain axis, offering a holistic approach to alleviating depression through microbial and neurophysiological synergy. Current evidence supports exercise as a potent adjunct therapy for depression, with personalized regimens (e.g., aerobic, resistance, or yoga) tailored to individual microbiota profiles. However, challenges remain in optimizing exercise prescriptions (intensity, duration, and type) and integrating them with probiotics, prebiotics, or FMT for synergistic effects. Future research should prioritize large-scale randomized controlled trials to validate causality, multi-omics approaches to decipher MGB axis dynamics, and mechanistic studies exploring microbial metabolites as therapeutic targets. The authors advocate for a paradigm shift toward microbiota-centric interventions, emphasizing the bidirectional relationship between physical activity and gut ecosystem resilience in mental health management. In conclusion, this review underscores exercise as a multifaceted modulator of the gut-brain axis, offering novel insights into non-pharmacological strategies for depression. By bridging microbial ecology, neuroimmunology, and exercise physiology, this work lays a foundation for precision medicine approaches targeting the gut microbiota to alleviate depressive disorders.

ObjectiveTo explore quality difference factors of Perillae Caulis based on the contents of multiple chemical components and comprehensively evaluate the quality. MethodsA total of 32 batches of Perillae Caulis samples were collected from 12 producing areas such as Hebei， Anhui and Guangdong， and their diameter range， epidermis color and producing areas were recorded. Total flavonoids， total phenols， volatile oils， 5 active components and 84 volatile components in 32 batches of samples were quantitatively or semi-quantitatively determined by colorimetry， ultra performance liquid chromatography-photodiode array detector（UPLC-PDA） and gas chromatography-mass spectrometry（GC-MS）. Then the differences between the contents of these components were analyzed by principal component analysis（PCA） and non-parametric test. According to the weights of the index components determined by PCA model， entropy weight-technique for order preference by similarity to ideal solution（TOPSIS） model was constructed to evaluate the quality of Perillae Caulis with different characters and origins. ResultsThere were significant differences in the composition of Perillae Caulis with different diameters， epidermis colors and producing areas， and 9 differential components were screened out， including 6 index constituents（total flavonoids， total phenols， caffeic acid， scutellarin， rosmarinic acid and luteolin） and 3 volatile components（caryophyllene oxide， （-）-humulene epoxide Ⅱ， 14-hydroxycaryophyllene）， of which 6 index constituents were higher in samples with small diameter， purple-brown epidermis and southern origin， while the contents of 3 volatile components were higher in samples with large diameter， dark-brown epidermis and northern origin. A significant difference was shown in the model scores of different diameters， epidermis colors and origins（P<0.05）， and the scores of Perillae Caulis with small diameter and purple-brown epidermis from southern area， especially Guangdong， had a high score. ConclusionThere are significant differences in the composition and content of chemical constituents between different diameters， epidermal colors and production areas of Perillae Caulis， samples showing small diameter， owing purple-brown epidermis， and originating from Guangdong were of higher-quality due to their higher content of 8 key indices.

ObjectiveTo explore quality difference factors of Perillae Caulis based on the contents of multiple chemical components and comprehensively evaluate the quality. MethodsA total of 32 batches of Perillae Caulis samples were collected from 12 producing areas such as Hebei， Anhui and Guangdong， and their diameter range， epidermis color and producing areas were recorded. Total flavonoids， total phenols， volatile oils， 5 active components and 84 volatile components in 32 batches of samples were quantitatively or semi-quantitatively determined by colorimetry， ultra performance liquid chromatography-photodiode array detector（UPLC-PDA） and gas chromatography-mass spectrometry（GC-MS）. Then the differences between the contents of these components were analyzed by principal component analysis（PCA） and non-parametric test. According to the weights of the index components determined by PCA model， entropy weight-technique for order preference by similarity to ideal solution（TOPSIS） model was constructed to evaluate the quality of Perillae Caulis with different characters and origins. ResultsThere were significant differences in the composition of Perillae Caulis with different diameters， epidermis colors and producing areas， and 9 differential components were screened out， including 6 index constituents（total flavonoids， total phenols， caffeic acid， scutellarin， rosmarinic acid and luteolin） and 3 volatile components（caryophyllene oxide， （-）-humulene epoxide Ⅱ， 14-hydroxycaryophyllene）， of which 6 index constituents were higher in samples with small diameter， purple-brown epidermis and southern origin， while the contents of 3 volatile components were higher in samples with large diameter， dark-brown epidermis and northern origin. A significant difference was shown in the model scores of different diameters， epidermis colors and origins（P<0.05）， and the scores of Perillae Caulis with small diameter and purple-brown epidermis from southern area， especially Guangdong， had a high score. ConclusionThere are significant differences in the composition and content of chemical constituents between different diameters， epidermal colors and production areas of Perillae Caulis， samples showing small diameter， owing purple-brown epidermis， and originating from Guangdong were of higher-quality due to their higher content of 8 key indices.

This study aims to explore the mechanism of Huanglian Jiedu Decoction(HJD) in treating acute liver injury(ALI) in the mouse model of sepsis induced by lipopolysaccharide(LPS). Fifty-four male C57BL/6 mice were randomized into six groups: blank group, model group, low-, medium-, and high-dose group HJD, and dexamethasone group. The mouse model of sepsis was established by intraperitoneal injection of LPS after 7 days of gavage with HJD, and dexamethasone(0.2 mL) was injected intraperitoneally 1.5 h after modeling. The murine sepsis score(MSS) was recorded 12 h after modeling. The levels of alanine aminotransferase(ALT) and aspartate aminotransferase(AST) in the liver tissue and tumor necrosis factor-α(TNF-α) and interleukin-6(IL-6) in the serum were measured by ELISA. Hematoxylin-eosin(HE) staining was used to observe the pathological changes of the mouse liver. The content of light chain 3 of microtubule-associated protein 1(LC3) was detected by immunofluorescence, and that of sirtuin 1(SIRT1) was detected by immunohistochemistry. The mRNA levels of adenosine 5'-monophosphate-activated protein kinase(AMPK), LC3, and P62 were detected by RT-PCR. Western blot was employed to determine the protein levels of AMPK, p-AMPK, and SIRT1 in the liver tissue. The results showed that compared with model group, drug interventions decreased the MSS and liver injury indicators, lowered the levels of inflammatory cytokines, improved the liver tissue structure, upregulated the protein levels of of p-AMPK/AMPK and SIRT1 and the mRNA levels of AMPK and LC3, and downregulated the mRNA level of P62. To sum up, HJD can regulate the autophagy level and reduce inflammation to ameliorate acute liver injury in septic mice by activating the AMPK/SIRT1 autophagy pathway.
Animals ; Drugs, Chinese Herbal/administration & dosage* ; Sirtuin 1/genetics* ; Male ; Mice ; Sepsis/metabolism* ; Mice, Inbred C57BL ; Autophagy/drug effects* ; AMP-Activated Protein Kinases/genetics* ; Liver/metabolism* ; Humans ; Signal Transduction/drug effects* ; Disease Models, Animal ; Tumor Necrosis Factor-alpha/genetics*

Objective:To study the relationship between serum adiponectin level and early vascular aging (EVA).Methods:The cross-sectional study method was used. Six hundred and seventy-two subjects who completed health checkup from June to December 2023 in the Affiliated Hospital of Qingdao University were selected. The subjects were divided into the EVA group (237 cases) and the control group (435 cases) based on brachial-ankle pulse wave velocity (baPWV). According to the adiponectin tertiles method, the subjects were divided into low adiponectin subgroup (2.4 to 6.6 mg/L, 225 cases), medium adiponectin subgroup (6.7 to 9.1 mg/L, 227 cases) and high adiponectin subgroup (9.2 to 19.8 mg/L, 220 cases). The basic demographic information, past history and serological indexes were recorded. Univariate and multivariate binary Logistic regression analyses were used to analyze the risk factors for EVA, and multivariate Logistic regression was used to analyze the effect of adiponectin on EVA.Results:The male proportion, age, body mass index (BMI), systolic blood pressure, diastolic blood pressure, triglycerides (TG), fasting blood glucose (FBG), uric acid, glycated hemoglobin (HbA 1c), homocysteine, baPWV and alcohol history proportion in EVA group were significantly higher than those in control group: 64.98% (154/237) vs. 53.33% (232/435), 53 (47, 57) years old vs. 46 (39, 52) years old, (26.34 ± 3.37) kg/m 2 vs. (25.16 ± 3.91) kg/m 2, (132.27 ± 15.48) mmHg (1 mmHg = 0.133 kPa) vs. (117.30 ± 13.04) mmHg, (81.79 ± 11.04) mmHg vs. (71.93 ± 10.10) mmHg, 1.45 (1.03, 2.03) mmol/L vs. 1.08 (0.76, 1.65) mmol/L, 5.52 (5.03, 6.21) mmol/L vs. 5.14 (4.77, 5.56) mmol/L, (380.04 ± 96.43) μmol/L vs. (362.18 ± 94.94) μmol/L, 5.80 (5.50, 5.90)% vs. 5.70 (5.40, 5.82)%, 10.70 (9.01, 12.90) μmol/L vs. 9.96 (8.30, 12.20) μmol/L, 1 586 (1 511, 1 719) cm/s vs. 1 299 (1 215, 1 367) cm/s and 19.41% (46/237) vs. 13.56% (59/435), the adiponectin was significantly lower than that in control group: 7.00 (5.70, 8.75) mg/L vs. 8.40 (6.40, 10.60) mg/L, and there were statistical differences ( P<0.01 or <0.05). There were no statistical differences in total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), creatinine and smoking history proportion between two groups ( P>0.05). The male proportion, BMI, systolic blood pressure, diastolic blood pressure, TG, FBG, uric acid, creatinine, HbA 1c, homocysteine, EVA incidence, baPWV, smoking history proportion and alcohol history proportion in low adiponectin subgroup and medium adiponectin subgroup were significantly higher than those in high adiponectin subgroup, furthermore, the indexes except HbA 1c in low adiponectin subgroup were significantly higher than those in medium adiponectin subgroup, and there were statistical differences ( P<0.05); the HDL-C in low adiponectin subgroup and medium adiponectin subgroup was significantly lower than that in high adiponectin subgroup, furthermore, that in low adiponectin subgroup was significantly lower than that in medium adiponectin subgroup, and there were statistical differences ( P<0.05); there were no statistical differences in age, TC and LDL-C among the three subgroups ( P>0.05). Univariate binary Logistic regression analysis result showed that age, male, BMI, alcohol history, systolic blood pressure, diastolic blood pressure, TG, FBG, uric acid and HbA 1c were the risk factors for EVA ( P<0.01 or <0.05), while the adiponectin was a protective factor for EVA ( P<0.01). Multivariate binary Logistic regression analysis result showed that age, systolic blood pressure, TG and FBG were risk factors for EVA ( OR = 1.098, 1.066, 1.209 and 1.268; 95% CI 1.069 to 1.127, 1.050 to 1.082, 1.007 to 1.451 and 1.069 to 1.502; P<0.01 or <0.05), while adiponectin was a protective factor ( OR = 0.892, 95% CI 0.828 to 0.962, P<0.01). Multivariable Logistic regression analysis result showed that adiponectin consistently remained a protective factor for EVA across unadjusted, preliminary adjusted and fully adjusted covariate models ( OR = 0.553, 0.580 and 0.576; 95% CI 0.451 to 0.678, 0.440 to 0.764 and 0.435 to 0.763; P<0.01). Conclusions:The serum APN level is negatively correlated with the risk of EVA, which may be an independent protective factor for the EVA.

Objective:To evaluate the clinical utility of an intelligent endoscope transportation system in the digestive endoscopy center.Methods:A parallel-group controlled trial was conducted at Digestive Endoscopy Center of Qilu Hospital of Shandong University from June 1st to December 31st 2024, comparing robotic intelligent endoscope transport (experimental group) versus manual transport (control group). Performance metrics, including response time, transportation speed, labor efficiency, contamination prevention, closed-loop traceability, and nursing staff satisfaction, were statistically analyzed. Full-time equivalent (FTE) was introduced to quantify the operational efficiency of the experimental group.Results:The study included a total of 60 206 instances of intelligent endoscope transportation and 60 485 instances of manual transportation data. The robotic group demonstrated significantly shorter response times versus manual group for initial dispatch (51.08±14.97 seconds VS 54.44±13.61 seconds, t=35.8, P<0.001) and recovery response time (32.52±11.26 seconds VS 40.20±11.40 seconds, t=103.93, P<0.001). During the 148 days operational period, the success rate was 99.83% (60 104/60 206) and the failure rate was 0.17% (102/60 206) for robotic transports. Primary failure causes were wireless disconnection, pathfinding errors, and mechanical faults, averaging 1.05 malfunctions/month with no adverse events. The success and failure rate was 99.26% (60 043/60 485) and 0.74% (442/60 485) respectively for manual transports. Staff satisfaction was significantly higher for robotic transport in endoscopic transportation (4.65±0.55 scores VS 3.97±0.98 scores, t=96.5, P<0.001) and delivery process (4.71±0.59 scores VS 3.90±1.04 scores, t=210.3, P<0.001). and workload intensity was significantly lower (4.06±0.77 scores VS 4.48±0.63 scores, t=59.9, P=0.025). The system reduced labor requirements by 3.68 FTE, yielding annual savings of ￥657 000. Conclusion:The robotic intelligent endoscope transport system improves work efficiency, reduces nursing labor costs and physical workload, enhances job experience and satisfaction, and enables full-process smart traceability, providing a validated solution for endoscopy center logistics.

Objective:To explore the association between circadian syndrome (CircS), metabolic syndrome (MetS) and mild cognitive impairment (MCI) in elderly rural adults in China.Methods:From March to September 2018, totally 5 765 participants aged 60 years or older from 52 villages in Yanlou Town, Yanggu County, Shandong Province were selected. The data included demographic, underlying disease and neuropsychological data were collected by questionaire survey. Having ≥3 of the following components was defined as MetS: elevated waist circumference, high triglycerides, low high-density lipoprotein cholesterol, elevated blood pressure and elevated fasting glucose. Having ≥4 of the following components was defined as CircS: short sleep (<6 h/d), depression and five other components which were used to define MetS, with elevated waist circumference as a mandatory item. MCI was diagnosed according to Petersen's criteria and further classified into amnestic MCI (aMCI) and non-amnestic MCI (naMCI) based on whether the memory domains impaired.Data were analyzed using multivariable Logistic regression and general linear regression models by R statistical software.Results:In the total sample ( n=4 898), 1 280 participants were diagnosed with MCI, of which 1 075 were aMCI and 205 were naMCI.Compared to the normal group, CircS alone was significantly associated with increased risks of MCI ( B=0.695, P=0.039) and aMCI ( B=0.782, P=0.024), as well as lower verbal fluency scores ( B=-0.244, P=0.045). No significant associations were found between MetS alone or both MetS and CircS and cognitive impairment( P>0.05). At the component level, short sleep and depression were associated with increased risks of MCI ( B=0.167, P=0.025; B=0.605, P<0.001) and aMCI ( B=0.185, P=0.020; B=0.600, P<0.001). Conclusion:Individuals with CircS are at a higher risk of cognitive impairment, CircS is more strongly associated with cognitive impairment than MetS, with short sleep duration and depressive symptoms potentially playing key roles.