Objective:To develop an artificial intelligence (AI)-based platelet review strategy to identify abnormal platelet histograms with no significant difference between initial impedance platelet count (PLT-I) and PLT-F results.Methods:This study included 5 119 routine blood analysis in Nanfang Hospital of Southern Medical University and its Ganzhou branch from July 2023 and March 2024. Specimens exhibiting abnormal platelet histograms and an initial platelet count >40×10?/L underwent review using the fluorescent platelet count (PLT-F) channel. Consistency of the results was defined as a difference between impedance platelet count (PLT-I) and PLT-F less than ±20% of the PLT-F results. A deep learning model was developed using platelet and red blood cell histogram data from a training set of 3 807 specimens. The model′s diagnostic performance was evaluated on an independent external validation set ( n=805) using receiver operating characteristic (ROC) curve analysis. Changes in the number of reviewed samples and sample turnaround time were analyzed to assess its clinical utility. Results:The deep learning model based on platelet and red blood cell histograms achieved an area under the ROC curve (AUC) of 0.854 in the training set. At a cutoff value of 0.1, the sensitivity was 0.954 and specificity was 0.358. The model could reduce review by 16.80% (190/1 131). In the validation set, the AUC was 0.805, with a sensitivity of 0.955 and specificity of 0.307, corresponding to a reduction of 17.41% (47/270) in reviewed specimens.Conclusion:The platelet review prediction model developed based on deep learning technology can efficiently identify samples with consistent results before and after review, reducing unnecessary reviews and shortening specimen testing time, thereby improving the efficiency of platelet test.

Objective:To experimentally validate clinical samples, analyze the mRNA expression of the FYVE domain containing phosphatidylinositol 3-phosphate 5 kinase ( PIKFYVE) gene, and its clinical significance based on the Cancer Genome Atlas (TCGA) database in hepatocellular carcinoma (HCC). Methods:Data information on 424 clinical samples (including 374 cases of HCC tissues and 50 cases of non-tumorous liver tissues) were collected based on the TCGA database. Cox regression analysis and the Kaplan-Meier method were used to analyze the relationship between mRNA expression of the PIKFYVE gene and the clinical characteristics as well as survival prognosis in patients with HCC. The relationship between the PIKFYVE gene and immune cell infiltration was examined by correlation analysis with 24 kinds of immune cells. In addition, the mRNA expression level of the PIKFYVE gene and RAC-alpha serine/threonine-protein kinase ( AKT1), phosphatase and tensin homolog ( PTEN), protein kinase C alpha ( PRKCA), inositol polyphosphate-5-phosphatase ( INPP5D), phosphoinositide-3-kinase regulatory subunit 1 ( PIK3R1), inositol polyphosphate 4-phosphatase type II ( INPP4B) and phospholipase C beta 4 ( PLCB4) gene correlations were analyzed in HCC tissues. At the same time, paraffin sections of highly differentiated, moderately differentiated, poorly differentiated, and non-tumor liver tissues from patients with HCC were collected from the Department of Pathology of the First Affiliated Hospital of Xinjiang Medical University. The histopathological observation was performed by HE staining. Immunohistochemistry was used to verify the expression levels of the PIKFYVE and Ki67 proteins in each clinical sample. The t-test was used for intergroup comparison of continuous data. The χ2 test and Wilcoxon rank sum test were used for intergroup comparison of enumeration data. The Kaplan-Meier method was used for survival analysis. Results:The expression level of the PIKFYVE gene was higher in the HCC tumor than that in normal liver tissue ( P<0.01). The overall survival time of patients was significantly longer in the low expression group than that in the high expression group ( HR=1.57, 95% CI: 1.10～2.25, P=0.014). The results of univariate Cox regression analysis showed that tumor stage, pathological grade, tumor status, residual tumor, and PIKFYVE expression level all had an effect on OS ( P<0.05). The PIKFYVE prognostic risk model had a proportionate score of HR=1.533 (95% CI: 1.077～2.181, P=0.018). Multivariate Cox risk regression analysis showed that the PIKFYVE prognostic risk model had a proportionate score of HR=1.481 (95% CI: 0.886～2.476, P=0.134) and an area under the receiver operating characteristic curve of 0.559, indicating that it had predictive value for survival prediction. The results of the correlation analysis showed that the expression level of PIKFYVE was strongly correlated with immune cell infiltration and TP53 ( P<0.01). The results of immunohistochemical staining showed that the expression level of PIKFYVE was significantly higher in HCC tissue samples than that in non-tumor liver tissues ( P<0.01), and was negatively correlated with the degree of differentiation. Conclusion:PIKFYVE, as an independent risk factor, is expected to be developed into a biomarker for clinical diagnosis, offering a reference for novel therapeutic agents in HCC.

Objective To Explore the effect of expression of peroxidase reductase 4(PRDX4)on cervi-cal lymph node metastasis in oral squamous cell carcinomas(OSCC)and its relationship with prognosis.Methods Using bioinformatics databases to explore the expression,prognosis analysis,and related pathway prediction of PRDX4 gene in head and neck squamous cell carcinoma.At the same time,clinical and pathologi-cal data including cervical lymph node metastasis were collected from 124 OSCC patients,the expression of PRDX4 was detected in 124 OSCC tissues using immunohistochemical experimental methods.Results Analy-sis of online bioinformatics databases showed that PRDX4 was highly expressed in head and neck squamous cell carcinoma and significantly correlated with tumor grade.The immunohistochemical results showed that the occurrence of cervical lymph node metastasis was significantly correlated with the intensity of PRDX4 stai-ning in 124 OSCC cases studied,and high expression of PRDX4 was an independent risk factor for cervical lymph node metastasis in OSCC(P=0.010).The staining intensity of PRDX4 was significantly correlated with the presence of multiple lymph node metastasis(P=0.020)and the the maximum diameter of lymph node metastasis(P=0.031),but not related to the invasion of lymph nodes outside the membrane.Survival a-nalysis showed that strong PRDX4 positivity was significantly correlated with poor disease-free survival(DFS)rate and overall survival(OS)rate in OSCC patients.Conclusion OSCC patients with strong expres-sion of PRDX4 have a higher incidence of cervical lymph node metastasis and a poorer prognosis.

Objective To investigate the correlation of sarcoplasmic/endoplasmic reticulum Ca2+ATPase 2a(SERCA2a)and blood urea nitrogen(BUN)to albumin(ALB)ratio(BAR)with cardiac function and prognosis in patients with chronic heart failure(CHF).Methods A total of 200 CHF patients admitted to this hospital from January 2021 to January 2023 were selected as the CHF group and 100 healthy people un-dergoing physical examination as the control group.CHF patients were divided into class Ⅰ group(38 cases),class Ⅱ group(40 cases),class Ⅲ group(54 cases)and class Ⅳ group(68 cases)according to the New York Heart Association(NYHA)cardiac function classification after admission.According to the prognosis,they were divided into poor prognosis group(60 cases)and good prognosis group(140 cases).Serum SERCA2a BUN,ALB levels were measured and BAR was calculated.Spearman rank correlation coefficient was used to analyze the correlation between serum SERCA2a,BAR and NYHA cardiac function classification in CHF pa-tients.Taking the prognosis of CHF patients as the dependent variable,a multivariate unconditional Logistic regression model was established to determine its influencing factors.Receiver operating characteristic curve was drawn to evaluate the predictive value of serum SERCA2a and BAR for poor prognosis of CHF patients.Results Compared with the control group,the serum SERCA2a level was decreased and BAR was increased in the CHF group(P＜0.05).The level of SERCA2a in class Ⅰ group,class Ⅱ group,class Ⅲ group and classⅣ group decreased and BAR increased successively(P＜0.05).The NYHA classification was negatively cor-related with serum SERCA2a level(rs=-0.756,P＜0.05),and positively correlated with BAR(rs=0.722,P＜0.05).After 1-year follow-up,the incidence of poor prognosis in 200 CHF patients was 30.00％(60/200).Increased NYHA class(OR=1.936,95％CI:1.115-3.364),atrial fibrillation(OR=3.269,95％CI:1.078-9.913),N-terminal pro-B-type natriuretic peptide(OR=1.002,95％CI:1.001-1.009)and BAR(OR=1.169,95％CI:1.082-1.263)were independent risk factors for poor prognosis in patients with CHF(P＜0.05),increased left ventricular ejection fraction(OR=0.810,95％CI:0.716-0.916)and SERCA2a(OR=0.964,95％CI:0.947-0.981)were independent protective factors(P＜0.05).The area under the curve of serum SERCA2a combined with BAR to predict the poor prognosis of CHF patients was 0.865(95％CI:0.810-0.910),which was larger than 0.784(95％CI:0.720-0.839)and 0.777(95％CI:0.713-0.833)predicted by serum SERCA2a and BAR levels alone,and the difference was statistically significant(Z=2.944,3.250,P＜0.05).Conclusion The decrease of SERCA2a and the increase of BAR are closely re-lated to cardiac function and prognosis in patients with CHF.Serum SERCA2a combined with BAR has a higher value in predicting poor prognosis in patients with CHF.

Objective To analyze the PLCβ4 gene mRNA expression and its clinical significance in hepatocellular carcinoma (HCC) based on TCGA database. Methods Based on the data on 424 clinical samples (including 374 cases of HCC tissues and 50 cases of nontumor liver tissues) in the TCGA database, Kaplan–Meier method, Cox regression analysis, and immune infiltration analysis were performed to evaluate the relationship between PLCβ4 gene and the clinical characteristics and survival prognosis of HCC patients. Correlation analysis between PLCβ4 gene and 24 types of immune cells was applied to investigate the relationship between PLCβ4 gene and immune cell infiltration and mRNA expression level of TP53 gene, a high-frequency mutation gene in HCC. In addition, paraffin sections of highly, moderately, and poorly differentiated tumor tissues and normal liver tissues from HCC patients were collected. The histopathological observation was carried out via HE staining method, and the expression levels of PLCβ4 and Ki-67 proteins in each clinical sample were verified through the immunohistochemical method. Results The expression level of PLCβ4 gene in HCC was significantly higher than that in normal tissues (P<0.01), and all patients in the PLCβ4 high-expression group had a significantly longer overall survival than those in the low-expression group (P<0.05), which suggested that PLCβ4 substantially affected the prognosis of HCC patients. Correlation analysis showed that the expression level of PLCβ4 gene was highly correlated with immune cell infiltration and the expression level of TP53 gene. As verified by clinical sample experiments, HE staining experiments and immunohistochemical results revealed that PLCβ4 gene expression in HCC tissue samples was significantly higher than that in normal tissues (P<0.001), and it was negatively correlated with the degree of differentiation. Conclusion PLCβ4 may serve as an independent prognostic factor in HCC and is expected to be a novel molecular target for HCC treatment.

AIM: To identify the primary active components and underlying mechanisms of Yijing Decoction(YJD)in treating early diabetic retinopathy(DR)based on liquid chromatography-mass spectrometry and network pharmacology.METHODS: Active components of YJD were characterized through LC-MS. Components with optimal ADME(absorption, distribution, metabolism, excretion)properties were selected as key bioactive candidates. Network pharmacology approaches were employed to predict YJD-DR therapeutic targets. Protein-protein interaction(PPI)networks, gene ontology(GO)enrichment analysis, and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway analysis were subsequently conducted to predict core targets and networks. Critical targets and pathways were experimentally validated through Western blot.RESULTS: Ten core therapeutic targets were identified, including TNF, Alb, EGFR, STAT3, PTGS2, ESR1, PPAR, MMP9, TLR4, and MAPK. YJD was related to cancer-related signaling, fluid shear stress and atherosclerosis, and neurodegenerative diseases, encompassing key biological processes such as inflammatory response regulation, programmed cell death activation, and enhanced cell migration. Furthermore, Western blot analysis confirmed that YJD significantly inhibited high glucose-induced phosphorylation of STAT3(P-STAT3/STAT3)and ERK(P-ERK/ERK)in rat retinal microvascular endothelial cells.CONCLUSION: This study revealed YJD's pharmacodynamical basis and its multi-component, multi-target, and multi-paths pharmacology. YJD exerts therapeutic effects on DR by coordinately regulating critical signaling pathways and alleviating intraocular inflammation, thus preserving retinal vascular endothelial cells, maintaining blood-retinal barrier integrity, and facilitating retinal neurovascular repair.

Objective:To develop an artificial intelligence (AI)-based platelet review strategy to identify abnormal platelet histograms with no significant difference between initial impedance platelet count (PLT-I) and PLT-F results.Methods:This study included 5 119 routine blood analysis in Nanfang Hospital of Southern Medical University and its Ganzhou branch from July 2023 and March 2024. Specimens exhibiting abnormal platelet histograms and an initial platelet count >40×10?/L underwent review using the fluorescent platelet count (PLT-F) channel. Consistency of the results was defined as a difference between impedance platelet count (PLT-I) and PLT-F less than ±20% of the PLT-F results. A deep learning model was developed using platelet and red blood cell histogram data from a training set of 3 807 specimens. The model′s diagnostic performance was evaluated on an independent external validation set ( n=805) using receiver operating characteristic (ROC) curve analysis. Changes in the number of reviewed samples and sample turnaround time were analyzed to assess its clinical utility. Results:The deep learning model based on platelet and red blood cell histograms achieved an area under the ROC curve (AUC) of 0.854 in the training set. At a cutoff value of 0.1, the sensitivity was 0.954 and specificity was 0.358. The model could reduce review by 16.80% (190/1 131). In the validation set, the AUC was 0.805, with a sensitivity of 0.955 and specificity of 0.307, corresponding to a reduction of 17.41% (47/270) in reviewed specimens.Conclusion:The platelet review prediction model developed based on deep learning technology can efficiently identify samples with consistent results before and after review, reducing unnecessary reviews and shortening specimen testing time, thereby improving the efficiency of platelet test.

Objective:To experimentally validate clinical samples, analyze the mRNA expression of the FYVE domain containing phosphatidylinositol 3-phosphate 5 kinase ( PIKFYVE) gene, and its clinical significance based on the Cancer Genome Atlas (TCGA) database in hepatocellular carcinoma (HCC). Methods:Data information on 424 clinical samples (including 374 cases of HCC tissues and 50 cases of non-tumorous liver tissues) were collected based on the TCGA database. Cox regression analysis and the Kaplan-Meier method were used to analyze the relationship between mRNA expression of the PIKFYVE gene and the clinical characteristics as well as survival prognosis in patients with HCC. The relationship between the PIKFYVE gene and immune cell infiltration was examined by correlation analysis with 24 kinds of immune cells. In addition, the mRNA expression level of the PIKFYVE gene and RAC-alpha serine/threonine-protein kinase ( AKT1), phosphatase and tensin homolog ( PTEN), protein kinase C alpha ( PRKCA), inositol polyphosphate-5-phosphatase ( INPP5D), phosphoinositide-3-kinase regulatory subunit 1 ( PIK3R1), inositol polyphosphate 4-phosphatase type II ( INPP4B) and phospholipase C beta 4 ( PLCB4) gene correlations were analyzed in HCC tissues. At the same time, paraffin sections of highly differentiated, moderately differentiated, poorly differentiated, and non-tumor liver tissues from patients with HCC were collected from the Department of Pathology of the First Affiliated Hospital of Xinjiang Medical University. The histopathological observation was performed by HE staining. Immunohistochemistry was used to verify the expression levels of the PIKFYVE and Ki67 proteins in each clinical sample. The t-test was used for intergroup comparison of continuous data. The χ2 test and Wilcoxon rank sum test were used for intergroup comparison of enumeration data. The Kaplan-Meier method was used for survival analysis. Results:The expression level of the PIKFYVE gene was higher in the HCC tumor than that in normal liver tissue ( P<0.01). The overall survival time of patients was significantly longer in the low expression group than that in the high expression group ( HR=1.57, 95% CI: 1.10～2.25, P=0.014). The results of univariate Cox regression analysis showed that tumor stage, pathological grade, tumor status, residual tumor, and PIKFYVE expression level all had an effect on OS ( P<0.05). The PIKFYVE prognostic risk model had a proportionate score of HR=1.533 (95% CI: 1.077～2.181, P=0.018). Multivariate Cox risk regression analysis showed that the PIKFYVE prognostic risk model had a proportionate score of HR=1.481 (95% CI: 0.886～2.476, P=0.134) and an area under the receiver operating characteristic curve of 0.559, indicating that it had predictive value for survival prediction. The results of the correlation analysis showed that the expression level of PIKFYVE was strongly correlated with immune cell infiltration and TP53 ( P<0.01). The results of immunohistochemical staining showed that the expression level of PIKFYVE was significantly higher in HCC tissue samples than that in non-tumor liver tissues ( P<0.01), and was negatively correlated with the degree of differentiation. Conclusion:PIKFYVE, as an independent risk factor, is expected to be developed into a biomarker for clinical diagnosis, offering a reference for novel therapeutic agents in HCC.

Objective:To investigate the intelligence status of children aged 10 - 12 in rural areas of Linxia Hui Autonomous Prefecture (referred to as Linxia) in Gansu Province.Methods:From September to November 2019, a cross-sectional study was carried out to investigate the intelligence status of children aged 10 to 12 in 8 counties (cities) of Linxia. Chinese Raven's Progressive Matrices (rural version) was used for intelligence test and children's intelligence quotient (IQ) was calculated by regular mold to evaluate children's intelligence level.Results:A total of 1 721 children in Linxia were tested for intelligence, with an average IQ of 103.2. Among them, low intelligence (≤69) accounted for 1.0% (18/1 721), borderline (70 - 79) accounted for 3.0% (52/1 721), middle and lower (80 - 89) accounted for 8.4% (144/1 721), moderate (90 - 109) accounted for 56.6% (974/1 721), middle and upper (110 - 119) accounted for 21.9% (377/1 721), excellent (120 - 129) accounted for 7.8% (135/1 721), extremely excellent (≥130) accounted for 1.2% (21/1 721). There were 635, 598 and 488 children aged 10, 11 and 12, respectively, with an average IQ of 106.1, 103.3 and 99.2. There were 919 males and 802 females, with an average IQ of 102.9 and 103.4, respectively.Conclusion:In 2019, the intelligence of children aged 10 to 12 in rural areas of Linxia has reached the moderate level.

Objective To analyze the predictive value of serum levels of procalcitonin(PCT)and cytokines on the prognosis of patients with COVID-19 at admission.Methods From November 2022 to February 2023,patients diagnosed with COVID-19 who were admitted to Beijing Chest Hospital were enrolled.Chemiluminescence was used to detect serum PCT levels,and flow microsphere array was used to detect serum cytokines IL-1β,IL-2,IL-4,IL-5,IL-6,IL-8,IL-10,IL-12p70,IL-17A,IL-17F,IL-22,TNF-α,TNF-β,IFN-γ level.ICU admission,mechanical ventilation and in-hospital death were defined as poor prognosis.After excluding patients with bacterial infection,the relationship between serum PCT and cytokine levels at admission and the prognosis of COVID-19 patients was analyzed.After excluding patients with bacterial infection,the relationship between serum PCT and cytokine levels at admission and the prognosis of COVID-19 patients was analyzed.Results A total of 176 patients with complete data were included,including 134 in the PCT-normal group and 42 in the PCT-elevated group,with a median age of 71.50 years and 71.59%males.Patients in the PCT elevated-group had significantly higher rates of ICU admission(38.41%vs.13.11%,P<0.05),mechanical ventilation(76.92%vs.24.59%,P<0.001)and in-hospital mortality(38.46%vs.6.56%,P<0.001)were significantly higher than those in the PCT-normal group.Serum levels of cytokines IL-6(7.40 pg/mL vs.4.78 pg/mL,P = 0.033 4)and IL-8(10.97 pg/mL vs.5.92 pg/mL,P<0.001)were significantly higher in patients with poor prognosis than in those with good prognosis.The area under the curve for PCT,IL-6,and IL-8 to predict poor prognosis in COVID-19 patients was 0.687,0.660,and 0.746,respectively;sensitivity was 52.78%,55.17%,and 72.41%,respectively;and specificity was 81.58%,74.19%,and 74.19%,respectively,as calculated from the ROC curve.When PCT,IL-6 and IL-8 jointly predict the prognosis of COVID-19 patients,the area under the curve is 0.764,the sensitivity is 70.00%,and the specificity is 80.00%.Conclusion Serum PCT and cytokines IL-6 and IL-8 could be used as predictive markers for poor prognosis in patients with COVID-19.