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 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 explore the effect of CT slice thickness on the reproducibility of radiomic features in urinary stones.Methods A retrospective analysis was performed on 30 cases of stones diagnosed by urinary CT.Thin-slice(1 mm)images were reconstructed into two groups of images with thicknesses of 2 mm and 5 mm.Two radiologists conducted manual segmentation independently and then extracted features.The reproducibility of the radiomic features was calculated using the concordance correlation coefficient(CCC).Results Under different slice thickness conditions by the same radiologist,the average CCC of all groups was less than 0.85(P＜0.05).For radiologist A,the proportion of stone features with CCC＜0.85 were 50.13％,79.91％,and 82.38％in the 1 mm vs 2 mm group,2 mm vs 5 mm group,and 1 mm vs 5 mm group,respectively.Corresponding values for radiologist B were 44.55％,79.47％and 82.32％,respectively.Among the seven categories of radiomic features,the morphological features with CCC＜0.85 was 100％in the 1 mm vs 5 mm group.Conclusion CT slice thickness significantly affected the reproducibility of radiomic features in urinary stones,with the greatest effect observed in morphological features.Thin-slice imaging demonstrated more stable reproducibility of stone characteristics.

Objective To establish the method of simultaneous determination of four main components of Danggui Liuhuang Decoction,including phellodendrine,palmatine,calycosin,and ferulic acid and provide reference for the quality control of Danggui Liuhuang Decoction.Methods Based on the HPLC-MS/MS analysis method,the positive ion data acquisition mode were adopted for the mass spectrometry detection and the four main components were quantified with multiple reaction monitoring mode(MRM)by ESI source.The chromatographic column was Agilent Extend-C18(5 μm,4.6 mm×250 mm),and gradient elution was performed with methanol and 0.5%formic acid in water.Results The linear range of phellodendrine was from 2-200 nmol/ml,and the linear range of palmatine,calycosin and ferulic acid was from 20-2 000 nmol/ml.The contents of the four components in the seven batches of Danggui Liuhuang Decoction were relatively stable,among which ferulic acid was mainly found in Phellodendrine and Coptidis;Phellodendrine was only detected in cortex phellodendri;the content of calycosin in Scutellaria baicalensis and Astragalus was higher;palmatine was detected in both Phellodendron and Astragalus.Conclusion The method had high sensitivity,good specificity and sample stability,which could meet the requirements of quantitative analysis of Traditional Chinese Medicine compounds,and could provide reference for further pharmacokinetics study on the content changes of traditional Chinese medicine compounds in biological samples.

Objective To investigate the value of thrombomodulin(TM),thrombin-antithrombin complex(TAT),α2 plasmin inhibitor-plasmin complex(PIC)and tissue plasminogen activator-inhibitor complex(t-PAIC)in the diagnosis and prognosis of neonatal disseminated intravascular coagulation(DIC).Methods Eighty-seven DIC neonates(the observation group)were included and divided into the survival group(66 cases)and the death group(21 cases)based on their outcomes at discharge.And 50 healthy newborns born in the same period were selected as the control group.The clinical data of neonates were collected,and risk factors of neonatal DIC were analyzed by Logistic regression.The differences of TM,TAT,PIC and t-PAIC levels in different groups were analyzed.The receiver operating characteristic(ROC)curve was used to analyze values of TM,TAT,PIC and t-PAIC in the diagnosis and prognosis of neonatal DIC.Results The incidence of low Apgar score,birth asphyxia,IVH,sepsis and maternal pregnancy induced hypertension syndrome(PIH)were higher in the observation group than those in the control group(P＜0.05).Multivariate Logistic regression analysis showed that low Apgar score,birth asphyxia,sepsis and PIH were independent risk factors for neonatal DIC.TM,TAT,PIC and t-PAIC levels were higher in the observation group than those in the control group(P＜0.05).ROC curve showed that the combined diagnosis value of TM,TAT,PIC and t-PAIC was better than that of single diagnosis of neonatal DIC.TM and TAT levels were higher in the death group than those in the survival group(P＜0.05),and there were no significant differences in PIC and t-PAIC levels between the two groups.Multivariate Logistic regression analysis showed that elevated TAT level was an independent risk factor for neonatal DIC prognosis.ROC curve showed that when TAT was 21.72 μg/L,the area under the curve for predicting neonatal DIC prognosis was 0.772(95%CI:0.666-0.878),and the sensitivity and specificity were 76.2%and 71.2%,respectively.Conclusion The combined application of TM,TAT,PIC and t-PAIC has important clinical value in diagnosis and prognosis evaluation of neonatal DIC.

Hyperhomocysteinemia (HHcy) is one of the independent risk factors for youth cerebral infarction. Gene mutation of key enzymes in homocysteine metabolism is the main cause of HHcy. Few cases of cystathionine beta-synthase (CBS) compound heterozygous mutation complicated with pulmonary embolism and lower extremity artery embolism have been reported. This article reported a young cerebral infarction patient complicated with pulmonary embolism and lower extremity artery embolism, who was subsequently detected with significantly elevated blood Hcy, and finally etiologically diagnosed with CBS 833 T>C/1082C>T compound heterozygous mutation. With the treatment of folic acid, methyl cobalt amine, vitamin B 6 and anticoagulant, the blood Hcy has been gradually declined, and no new thrombotic events occurred during the follow-up period of a year.

In order to better evaluate the transport effect of nanoparticles through the nasal mucosa, an nasal cavity-mimic model was designed based on M cells. The differentiation of M cells was induced by co-culture of Calu-3 and Raji cells in invert model. The ZO-1 protein staining and the transport of fluorescein sodium and dexamethasone showed that the inverted co-culture model formed a dense monolayer and possessed the transport ability. The differentiation of M cells was observed by up-regulated expression of Sialyl Lewis A antigen (SLAA) and integrin 1, and down-regulated activity of alkaline phosphatase. After targeting M cells with iRGD peptide (cRGDKGPDC), the transport of nanoparticles increased. , the co-administration of iRGD could result in the increase of nanoparticles transported to the brain through the nasal cavity after intranasal administration. In the evaluation of immune effect , the nasal administration of OVA-PLGA/iRGD led to more release of IgG, IFN-, IL-2 and secretory IgA (sIgA) compared with OVA@PLGA group. Collectively, the study constructed M cell model, and proved the enhanced effect of targeting towards M cell with iRGD on improving nasal immunity.

Objective To investigate whether vascular endothelial cells in choroidal neovascularization whether hypoxia condition can up-regulate SNAI1 and activate matrix metalloproteinase (MMP)2 and MMP9 therefore to participate in choroidal neovascularization(CNV).Methods Sixteen SPF male C57 mice aged 6-8 weeks were divided into control group and model group.CNV models were induced by retinal laser photocoagulation,and flatmount and frozen sections of retinal pigment epithelium (RPE)-choroid-sclera compound were prepared at 7 days after modeling.The CNV in flat-mount was examined by Isolectin B4 staining,and the location of SNAI1,MMP2 and MMP9 in frozen sections was determined by immunofluorescence technology.The expression of SNAI1,MMP2 and MMP9 at mRNA level in CNV was detected by real-time fluorescence quantitative PCR (real-time PCR).The use and care of experimental animals complied with Statement for the Use of Animals in Ophthalmic and Visual Research.The RF/6A cells were divided into normoxia group and hypoxia group and cultured for 24 hours in 5％ CO2condition and mix condition of 94％ N2,5％ CO2 and 1％ O2,respectively.The expression of SNAI1,MMP2 and MMP9 in the cells at mRNA and protein levels was detected by real-time PCR and Western blot assay,respectively.Small interfering RNA of SNAI1 (siSNAI1) was transfected into the cells,and then the expression of MMP2 in the cells at mRNA and protein levels was detected by real-time PCR and Western blot assay,respectively,and the migrating number of the cells was assayed by Transwell chamber assay.Results CD31 and SNAI1 positive-response cells were seen in RPE-choroid-sclera flat-mounts under the laser scanning confocal microscope.The relative expression levels of SNAI1mRNA and MMP2 mRNA in RPE-choroid-sclera tissues were higher in the model group than those in the control group (SNAI1 mRNA:1.291 ±0.060 vs.0.759±0.074,P =0.001;MMP2 mRNA:1.610±0.424 vs.0.772 ±0.080,P =0.044).The expression of MMP9 mRNA was not significantly elevated between model group and control group (P＞0.05).The relative expression level of MMP2 mRNA was higher in comparison with MMP9 mRNA in the model group (P＜0.01).The relative expressions of hypoxic induced factor 1α (HIF-1α),SNAI1 and MMP2 at mRNA level and protein level in RF/6A cells were significantly higher in the hypoxia group than those in the normoxia group (all at P＜0.05) and no considerable difference was seen in MMP9 mRNA expression between the two groups (P＞0.05).The relative expressions of MMP2 mRNA in the cells were 0.217±0.036 and 0.818±0.105,and those of MMP2 protein in the cells were 0.236±0.009 and 1.043±0.120 in the hypoxia+siSNAI1 group and only hypoxia group,respectively,with significant differences between them (P =0.002,0.003).The migrating number of the cells was (254.60 ±71.31)/field in the hypoxia+siSNAI1 group,which was significantly less than (534.10±96.21) /field in the control group (P =0.029).Conclusions The hypoxic environment at CNV can activate MMP2 by up-regulating the expression of SNAI1,which promotes the migration of vascular endothelial cells and therefore participates in CNV formation,and the intervention of SNAI1 activation under the hypoxic condition can inhibit this process.

Objectives: To investigate the expression of phosphoserine aminotransferase 1 (PSAT1) in pancreatic cancer tissues, and its potential role in pancreatic cancer. Methods: The expression of PSAT1 in 98 human pancreatic cancer tissues, which were collected from the People＇s Hospital of Guizhou, between July 2013 to July 2017, and the corresponding adjacent normal tissues was analyzed by immunohistochemical staining. Additionally, the relationship between the expression of PSAT1 and the clinicopathological parame-ters, overall survival (OS), and disease-free survival (DFS) of patients with pancreatic cancer was evaluated. The human pancreatic can-cer cell lines, BxPC-3 and SW1990, were transfected with PSAT1-siRNA, to investigate the effect of PSAT1 knockdown on cell prolifera-tion, migration, and invasion. Additionally, we performed Western blot to assess the expression of PI3K/Akt/mTOR-related proteins in PSAT1-knockdown cells. Results: The percentages of PSAT1-positive cells in pancreatic cancer and adjacent non-tumor tissues were 69.4% (68/98) and 5.0% (5/98), respectively, indicating a significantly higher expression of PSAT1 in pancreatic cancer tissues com-pared to adjacent non-tumor tissues (P<0.05). The increased expression of PSAT1 in pancreatic cancer tissues correlated with lymph node metastasis and TNM stage. Kaplan-Meier analysis suggested that a high expression of PSAT1 correlated with a poor OS and DFS compared to a low expression of PSAT1 (P<0.05). Cox regression analysis showed that the expression of PSAT1 is an independent prog-nostic marker for OS and DFS in pancreatic cancer patients (P<0.05, all). Transient transfection of BxPC-3 and SW1990 cells with PSAT1-siRNA markedly reduced the cell proliferation, migration, and invasion abilities of these cells compared to transfection with NC-siRNA (P<0.05). Knockdown of PSAT1 in pancreatic cancer cells also inhibited the expression of p-Akt and p-mTOR (P<0.05). Conclusions: The expression of PSAT1 increases in human pancreatic cancer tissues and cell lines. Additionally, PSAT1 regulates cell proliferation and in-vasion through the PI3K/Akt/mTOR pathway.

Objective To evaluate the effectiveness and prospect of innovative interactive follow-up mode for rehabilitation ofpostopera-tive breast cancer patietns in new medical environment. Methods The novel patient-centered medical care was promoted,and Pink Ribbon Clubs was established in different regions. Meanwhile,the patients’ self-support groups were formed, and a specialist for out-patients follow-up of outpatients by telephone or on the internet was arranged to provide psychological assessment,with timely and effective response. All data concerning to the follow-up were recorded and the patients were required to complete the relevant questionnaires. Results Compared with the same period in 2013,the follow-up rate was significantly increased to 80. 25%,and the patients showed higher satisfaction. The incidence of lymphedema in the patients was significantly reduced from 13% in 2013 to 7%. The quality of life was significantly improved and self-rat-ing score on depression was decreased from (51. 90 ± 8. 55) to (35. 81 ± 6. 21). Conclusion The kind of“Trinity” interactive mode is an effective follow-up mode with the clinical benefit in patients’ rehabilitation of physical and mental health,which is worthy of further applica-tion.