1.Establishment and Preliminary Analysis of an AG6 Mouse Encephalopathy Model Induced by Vaccinia Virus Tiantan Strain Infection
Lin YANG ; Meng JIN ; Hanqing WU ; Shun LI ; Xiaohui ZHOU
Laboratory Animal and Comparative Medicine 2026;46(1):3-10
ObjectiveA mouse model of vaccinia virus Tiantan strain (VTT)-induced encephalopathy was developed using AG6 mice. MethodsVTT was amplified by infecting Vero cells at a multiplicity of infection (MOI) of 0.01, followed by concentration and titration. After 72 h of incubation, virus-containing cells were collected and subjected to concentration. The concentrated viral suspension was serially diluted (10-fold dilutions) and added to 6-well plates containing confluent Vero cell monolayers for plaque assay. The number of plaques formed in each well was counted, and the virus titer was calculated based on the dilution factor. Fourteen 5-6-week-old AG6 mice (half male and half female, housed separately by sex) were randomly divided into a control group (n=3, PBS), a low-dose group (n=6, 1×10⁵ PFU), and a high-dose group (n=5, 5×10⁵ PFU). The mice were anesthetized by isoflurane inhalation and then infected via intranasal instillation. The mental state of the mice in each group was observed daily, and the body weight and mortality were recorded. On day 13 post-infection, 2% Evans Blue (4 mL/kg body weight) was administered via tail vein injection to assess blood-brain barrier (BBB) disruption. Subsequently, brain tissue samples were collected for immunofluorescence analysis to evaluate the activation of astrocytes and microglia. ResultsThe titer of purified VTT was 1×10⁷ PFU/mL. Compared with the control group, mice in the low-dose group showed no significant change in body weight, and no lethality was observed. In contrast, mice in the high-dose group exhibited significant weight loss starting on day 5 post-infection (P<0.05), accompanied by lethality. On day 13 post-infection, no Evans Blue extravasation was detected in the brain tissues of the low-dose group, while the olfactory bulb region of the high-dose group displayed distinct blue staining, indicating disruption of the BBB. Immunofluorescence analysis revealed no significant proliferation of astrocytes and microglia in the olfactory bulb region of the low-dose group on day 13 post-infection. In contrast, marked activation of glial cells was observable in the high-dose group. ConclusionAn animal model of VTT-induced encephalopathy in AG6 mice is successfully established, characterized by BBB disruption and reactive gliosis specifically localized to the olfactory bulb region, manifested as astrocytic and microglial proliferation.
2.Analysis of Variety Characteristics and Patterns of Marketed Traditional Chinese Patent Medicines for Treating Chronic Gastritis
Daiyue DING ; Changyue SONG ; Shuangfei DENG ; Siyu LI ; Xiangying KONG ; Xiaohui SU ; Na LIN
Chinese Journal of Experimental Traditional Medical Formulae 2026;32(11):252-260
ObjectiveThis study aims to systematically review the marketed traditional Chinese patent medicines for treating chronic gastritis (CG) in China. By analyzing their variety characteristics and prescription patterns, it seeks to provide references for clinical syndrome differentiation-based drug selection, treatment method optimization, and the design of high-quality clinical research. MethodsInformation on marketed traditional Chinese patent medicines for treating CG was collected. Microsoft Excel software was used to collate and statistically analyze representative drugs for each pathological stage, market status, syndrome types, and other contents. The Ancient and Modern Medical Case Cloud Platform (V2.3.9) was employed to analyze the formula composition patterns of standardized prescriptions. ResultsA total of 141 marketed traditional Chinese patent medicines for treating CG in China were included. Based on the disease's pathological progression, they can be classified into drugs for non-atrophic gastritis, atrophic gastritis, and precancerous lesions. Post-marketing research reveals that relevant evaluation is only conducted on 17 drugs, of which 2 involve pharmacoeconomic studies and 14 possess standardized evidence-based evidence. The primary dosage forms were capsules, granules, and tablets. From the 100 prescriptions screened according to inclusion/exclusion criteria, the varieties indicated for the stomach collateral stasis syndrome in atrophic gastritis accounted for the highest proportion. The main efficacy distributions were clearing heat, detoxifying, and relieving pain by promoting Qi circulation. Core drugs included Glycyrrhizae Radix et Rhizome, Paeoniae Radix Alba, and Aucklandiae Radix. Medicinal properties were predominantly warm and neutral. Flavors were mainly bitter, pungent, and sweet. The drugs primarily entered the spleen and stomach meridians. Analysis of the package inserts reveals that 67 products list "contraindications", 110 include "precautions", and 23 explicitly state "adverse reactions". ConclusionTraditional Chinese patent medicines for treating CG hold unique value in clinical practice. However, currently there are challenges such as insufficient clarity in syndrome type descriptions within package inserts and a relative lack of high-level evidence-based medical evidence, as well as pharmacoeconomic evaluations. Future efforts should focus on addressing these shortcomings by advancing research on syndrome characteristics and medication patterns based on syndrome differentiation, systematically conducting pharmacoeconomic evaluations, strengthening the accumulation of high-level evidence-based evidence, and, on this basis, improving patient medication adherence. This will comprehensively enhance the clinical application value and scientific connotation of this category of drugs.
3.Expert consensus on neoadjuvant PD-1 inhibitors for locally advanced oral squamous cell carcinoma (2026)
LI Jinsong ; LIAO Guiqing ; LI Longjiang ; ZHANG Chenping ; SHANG Chenping ; ZHANG Jie ; ZHONG Laiping ; LIU Bing ; CHEN Gang ; WEI Jianhua ; JI Tong ; LI Chunjie ; LIN Lisong ; REN Guoxin ; LI Yi ; SHANG Wei ; HAN Bing ; JIANG Canhua ; ZHANG Sheng ; SONG Ming ; LIU Xuekui ; WANG Anxun ; LIU Shuguang ; CHEN Zhanhong ; WANG Youyuan ; LIN Zhaoyu ; LI Haigang ; DUAN Xiaohui ; YE Ling ; ZHENG Jun ; WANG Jun ; LV Xiaozhi ; ZHU Lijun ; CAO Haotian
Journal of Prevention and Treatment for Stomatological Diseases 2026;34(2):105-118
Oral squamous cell carcinoma (OSCC) is a common head and neck malignancy. Approximately 50% to 60% of patients with OSCC are diagnosed at a locally advanced stage (clinical staging III-IVa). Even with comprehensive and sequential treatment primarily based on surgery, the 5-year overall survival rate remains below 50%, and patients often suffer from postoperative functional impairments such as difficulties with speaking and swallowing. Programmed death receptor-1 (PD-1) inhibitors are increasingly used in the neoadjuvant treatment of locally advanced OSCC and have shown encouraging efficacy. However, clinical practice still faces key challenges, including the definition of indications, optimization of combination regimens, and standards for efficacy evaluation. Based on the latest research advances worldwide and the clinical experience of the expert group, this expert consensus systematically evaluates the application of PD-1 inhibitors in the neoadjuvant treatment of locally advanced OSCC, covering combination strategies, treatment cycles and surgical timing, efficacy assessment, use of biomarkers, management of special populations and immune related adverse events, principles for immunotherapy rechallenge, and function preservation strategies. After multiple rounds of panel discussion and through anonymous voting using the Delphi method, the following consensus statements have been formulated: 1) Neoadjuvant therapy with PD-1 inhibitors can be used preoperatively in patients with locally advanced OSCC. The preferred regimen is a PD-1 inhibitor combined with platinum based chemotherapy, administered for 2-3 cycles. 2) During the efficacy evaluation of neoadjuvant therapy, radiographic assessment should follow the dual criteria of Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 and immune RECIST (iRECIST). After surgery, systematic pathological evaluation of both the primary lesion and regional lymph nodes is required. For combination chemotherapy regimens, PD-L1 expression and combined positive score need not be used as mandatory inclusion or exclusion criteria. 3) For special populations such as the elderly (≥ 70 years), individuals with stable HIV viral load, and carriers of chronic HBV/HCV, PD-1 inhibitors may be used cautiously under the guidance of a multidisciplinary team (MDT), with close monitoring for adverse events. 4) For patients with a poor response to neoadjuvant therapy, continuation of the original treatment regimen is not recommended; the subsequent treatment plan should be adjusted promptly after MDT assessment. Organ transplant recipients and patients with active autoimmune diseases are not recommended to receive neoadjuvant PD-1 inhibitor therapy due to the high risk of immune related activation. Rechallenge is generally not advised for patients who have experienced high risk immune related adverse events such as immune mediated myocarditis, neurotoxicity, or pneumonitis. 5) For patients with a good pathological response, individualized de escalation surgery and function preservation strategies can be explored. This consensus aims to promote the standardized, safe, and precise application of neoadjuvant PD-1 inhibitor strategies in the management of locally advanced OSCC patients.
4.Material Basis of Entada phaseoloides Against Neuropathic Pain Based on in vitro Chemical Profiling and in vivo Absorbed Constituents
Sensen LI ; Hanxue WU ; Yuanbin LI ; Limin ZHAO ; Xiaohui SU ; Hui XIONG ; Weihua MA ; Chao WANG ; Zhaochen MA ; Na LIN
Chinese Journal of Experimental Traditional Medical Formulae 2026;32(18):88-97
ObjectiveTo systematically characterize the in vitro chemical constituents of Entada phaseoloides and its absorbed prototype compounds and metabolites in a neuropathic pain (NP) model, thereby elucidating the potential pharmacological material basis underlying its analgesic activity. MethodsUltra-high-performance liquid chromatography coupled with quadrupole-Orbitrap high-resolution mass spectrometry (UHPLC-Q-Orbitrap HRMS) was used to analyze the constituents of the 80% methanol extract of E. phaseoloides. Thirty-six mice were divided into a blank group, low-, medium-, and high-dose E. phaseoloides extract groups (25, 50, and 100 mg·kg-1, respectively), an aspirin group (400 mg·kg-1), and a morphine group (10 mg·kg-1), with six mice in each group. The tail-flick threshold was measured. Twenty-four mice were randomly divided into a blank group and low-, medium-, and high-dose E. phaseoloides extract groups (25, 50, and 100 mg·kg-1, respectively), with six mice in each group, and motor coordination was evaluated. A spinal nerve ligation (SNL) model was established in mice. Twenty-four mice were randomly divided into a sham-operated (Sham) group, an SNL group, an SNL + E. phaseoloides extract group (750 mg·kg-1), and an SNL + pregabalin group (20 mg·kg-1), with 6 mice in each group. Mechanical allodynia and thermal hyperalgesia were evaluated. Behavioral assessments were combined to determine the optimal onset time of the analgesic effect, and plasma samples were collected at this time point. Absorbed components and metabolites were systematically identified based on accurate mass, MS/MS fragmentation patterns, retention time, and comparisons with literature data and reference standards. Meanwhile, network pharmacology, target similarity analysis, and molecular docking validation were integrated to evaluate the potential effects of absorbed components on disease intervention. ResultsCompared with the blank group, the high-dose E. phaseoloides extract group and morphine group showed significantly increased pain response thresholds (P<0.01), without affecting motor coordination in normal mice. Compared with the Sham group, the SNL group exhibited significantly decreased thresholds for mechanical allodynia and thermal hyperalgesia (P<0.01). Compared with the SNL group, the SNL + E. phaseoloides extract group and SNL + pregabalin group showed significantly increased thresholds for mechanical allodynia and thermal hyperalgesia (P<0.01). The analgesic effect of E. phaseoloides was most significant at 2 h after administration. A total of 70 constituents were identified in vitro, and 22 prototype compounds and 29 metabolites were detected in plasma. The targets of the absorbed components were mainly enriched in biological processes associated with pain signal perception and transmission and showed a high degree of overlap with disease genes related to neuropathic pain. In addition, multiple components exhibited strong binding energies with core pain-related targets. Based on integrated analysis of in vitro content, plasma exposure levels, and target associations, ketanthamide A-β-D-pyranoside and ketengzide were identified as the most representative pharmacologically active substances responsible for the analgesic effect. ConclusionThis study systematically elucidates the material basis of E. phaseoloides intervention in neuropathic pain from three aspects, including chemical constituent characteristics, in vivo behavioral characteristics, and potential targets, providing important theoretical foundations and experimental support for further studies on its pharmacological mechanism and analgesic drug development.
5.Mechanism of Qingrun Prescription-containing Serum Improving Insulin Resistance in HepG2 Cells via Branched-chain α-keto Acid Dehydrogenase Regulation of Branched-chain Amino Acids (BCAAs)/mTOR Pathway
Xiangwei BU ; Xiaohui HAO ; Runyun ZHANG ; Meizhen ZHANG ; Ze WANG ; Haoshuo WANG ; Jie WANG ; Qing NI ; Lan LIN
Chinese Journal of Experimental Traditional Medical Formulae 2025;31(9):90-98
ObjectiveTo investigate the effect of Qingrun prescription(QRP)-containing serum on improving insulin resistance in HepG2 cells and its potential mechanisms. MethodsAn insulin resistance model was established in HepG2 cells with 1×10-6 mol·L-1 insulin. Branched-chain α-keto acid dehydrogenase (BCKDH) gene silencing was achieved using siRNA, and the cells were divided into 8 groups: normal group, model group (1×10-6 mol·L-1 insulin), metformin group (1 mmol·L-1 metformin), high-, medium-, and low-dose QRP groups (20%, 10%, and 5% QRP-containing serum, respectively), QRP + siRNA-silenced BCKDH (si-BCKDH) group (10% QRP-containing serum + si-BCKDH), and QRP + si-NC group (10% QRP-containing serum + si-NC). Glucose levels in the supernatant were measured with a glucose assay kit, while glycogen content was assessed using a glycogen assay kit. Levels of branched-chain amino acids (BCAAs) and branched-chain keto acids (BCKAs) were determined using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). mRNA transcription and protein expression levels of BCKDH, dishevelled, Egl-10, and pleckstrin (DEP) domain-containing mammalian target of rapamycin (mTOR)-interacting protein (DEPTOR), mTOR, and ribosomal protein S6 kinase 1 (S6K1) were detected using real-time quantitative polymerase chain reaction (Real-time PCR) and Western blot. ResultsCompared to the normal group, the model group exhibited significantly decreased glucose consumption and glycogen content, increased levels of BCAAs and BCKAs, downregulated expression of BCKDH and DEPTOR, and upregulated mTOR and S6K1 expression (P<0.01). In comparison to the model group, QRP treatment at all doses significantly enhanced glucose consumption and glycogen content while reducing BCAAs and BCKAs levels (P<0.01). The high- and medium-dose QRP groups demonstrated significant upregulation of BCKDH mRNA transcription and protein expression, as well as DEPTOR mRNA transcription. Moreover, the DEPTOR protein expression level was significantly increased in high-, medium-, and low-dose QRP groups, while mTOR and S6K1 mRNA and protein expression levels were markedly downregulated (P<0.05, P<0.01). Compared to the QRP + si-NC group, the QRP + si-BCKDH group exhibited increased BCAAs and BCKAs levels, significantly decreased BCKDH mRNA transcription and protein expression, downregulated DEPTOR mRNA and protein expression, and upregulated mTOR and S6K1 mRNA and protein expression (P<0.05, P<0.01). ConclusionQRP may improve insulin resistance by reprogramming BCAAs metabolism. This effect involves upregulating BCKDH, reducing BCAAs and BCKAs levels, and suppressing the mTOR pathway activation.
6.Investigation of tick species in Suizhou City, Hubei Province from 2023 to 2024
Huiya LU ; Fang GUO ; Yibin PAN ; Meng PENG ; Libang WU ; Ye LIN ; Xiaohui LIU ; Xuejie YU
Chinese Journal of Schistosomiasis Control 2025;37(2):184-189
Objective To investigate the species of ticks in Suizhou City, Hubei Province, so as to provide insights into management of ticks and tick-borne diseases. Methods During the period between May 2023 and June 2024, livestock breeding farms and vegetation neighboring the place of residence of confirmed and suspected patients with tick-borne disease were selected as sampling points in rural areas from Yindian Township, Gaocheng Township, Wanhe Township, Wushan Township, Xiaolin Township, Xihe Township, Hedian Township and Beijiao Street in Suizhou City, Hubei Province, where confirmed and suspected cases with tick-borne diseases had been reported. The parasitic ticks on the body surface of free-range livestock were captured with tweezers in livestock breeding farms, and free ticks on the vegetation surface were captured with the flagging method. Morphological identification of tick samples was performed under a microscope, and the gender and developmental stage of ticks were determined. One engorged adult tick, 2 to 3 blood-feeding but non-engorged adult ticks, 10 to 15 unfed female ticks, 15 to 20 unfed male ticks, and 30 to 40 tick nymphs or larvae were assigned into a group, respectively. Genomic DNA was extracted from tick samples in each group, and mitochondrial 16S rRNA gene was amplified. Sequence analysis was performed with the DNASTAR software, and phylogenetic analysis was performed using the software MEGA 7.0. In addition, the phylogenetic tree was generated using the maximum likelihood method based on the Kimura 2 parameter model. Results A total of 2 438 ticks were captured from Suizhou City, Hubei Province during the period between May 2023 and June 2024, including 595 free ticks and 1 483 parasitic ticks. Three developmental stages of ticks were captured, including larvae, nymphs, and adults, and 75.18% (1 899/2 438) of captured ticks were adult, in which 79.04% (1 501/1 899) were female. Morphological and molecular biological assays identified one family, three genera and four species of captured ticks, including 2 425 Haemaphysalis longicornis ticks (99.47%) and one H. flava tick (0.04%) of the genus Haemaphysalis, 11 Rhipicephalus microplus ticks (0.45%) of the genus Rhipicephalus, and one Ixodes sinensis tick (0.04%) of the genus Ixodes in the family Ixodidae. Phylogenetic analysis revealed that the H. longicornis sequence (SZ49) in this study was clustered with sequences from Yunnan Province (GenBank accession number: MH024510.1), Hebei Province (GenBank accession number: MK450606.1) and Henan Province (GenBank accession number: MZ230645.1) into a clade, and the H. flava sequence (SZ19) in this study was clustered with sequences from Japan (GenBank accession number: MW064044.1), South Korea (GenBank accession number: ON629585.1), and Jiangsu Province (GenBank accession number: PP494741.1) and Hebei Province of China (GenBank accession number: MH520685.1) into a clade, while the R. microplus sequence (SZ8) in this study was clustered with the sequences from India (GenBank accession number: MK621328.1), and Henan Province (GenBank accession number: MT555307.1) and Guizhou Province of China (GenBank accession number: PP446801.1) into a clade. The sequence of I. sinensis (SZ23) in this study had 99.51% homology with that (GenBank accession number: OM368265.1) of ticks sampled from Wuhan City, Hubei Province. Conclusion There are four tick species of H. longicornis, H. flava, R. microplus and I. sinensis in Suizhou City, Hubei Province, and H. longicornis is the dominant species. H. flava is firstly recorded in Suizhou City.
7.Risk assessment of human Spirometra mansoni infections and cross-sectional study on knowledge, attitude and practice towards sparganosis in endemic areas of Henan Province
Yalan ZHANG ; Tiantian JIANG ; Xiaohui MA ; Yan DENG ; Weiqi CHEN ; Yankun ZHU ; Zhenqiang TANG ; Xi-meng LIN ; Hongwei ZHANG
Chinese Journal of Schistosomiasis Control 2025;37(2):190-195
Objective To assess the risk of human Spirometra mansoni infections and investigate the knowledge, attitude and practice (KAP) towards sparganosis mansoni among residents in Henan Province, so as to provide insights into formulation of the sparganosis mansoni control measures. Methods Qinling Village in Fugou County of Zhoukou City, Bali Village in Yancheng District of Luohe City, Duzhai Village in Puyang County of Puyang City and Doushan Village in Luoshan County of Xinyang City were sampled as survey sites in Henan Province from July to August 2023, and more than 40 frogs were sampled from ponds or streams in each survey site for detection of Sparganum mansoni infections. At least 150 residents were sampled using a cluster sampling method from each survey site, and the sero-prevalence of anti-S. mansoni IgG antibody was estimated. In addition, a questionnaire survey was conducted on the KAP towards sparganosis mansoni among participants, and the proportion of eligible KAP, rate of correct KAP and KAP scores were calculated. Results A total 229 frogs were collected from 4 survey sites in 2023, and the overall prevalence of S. mansoni infection was 4.37% (10/229) in frogs, with 7.75% (10/129) prevalence in wild frogs and 0 in farm-bred frogs. A questionnaire survey was performed among 649 residents sampled from 4 survey sites, and 649 serum samples were collected. The seroprevalence of anti-S.mansoni IgG antibody was 0.15% (1/649) and the overall proportion of eligible KAP was 23.73% (154/649) among participants. There were age- (χ2 = 30.905, P = 0.000), educational level- (χ2 = 41.011, P = 0.000), and occupation-specific proportions of eligible KAP among participants (χ2 = 10.721, P = 0.005), and the proportion of eligible KAP decreased with age (χ2 trend = 22.717, P = 0.000) and increased with education levels (χ2 trend = 40.025, P = 0.000). The rates of correct KAP towards sparganosis mansoni were 40.81% (2 119/5 192), 96.66% (1 882/1 947) and 63.81% (3 727/5 841) (χ2 = 1 913.731, P = 0.000) among residents, respectively. The rates of correct KAP towards sparganosis mansoni varied significantly among survey sites (χ2 = 136.872, 42.347 and 255.157; all P values= 0.000, with the highest rate of correct knowledge (51.94%, 748/1 440) and practices (75.86%, 1 229/1 620) in Yancheng District of Luohe City and the highest rate of correct attitudes in Puyang County of Puyang City (99.11%, 446/450) (all P values< 0.05). Conclusions There is still a high transmission risk of sparganosis mansoni in Henan Province, and the KAP towards sparganosis mansoni is required to be improved among residents.
8.Role and mechanism of biomimetic remineralization therapy for early enamel demineralization
Xiaohui LIN ; Mengyuan YANG ; Chunnian LI
Chinese Journal of Tissue Engineering Research 2025;29(4):856-865
BACKGROUND:With the improvement of diet and living standards,acidic diet and orthodontic treatment have become the main causes of enamel surface demineralization.As the first step of dental caries,enamel demineralization should be actively intervened.Mechanical grinding has great damage and does not conform to the concept of minimally invasive medicine.Biomimetic remineralization is the best way to deal with enamel demineralization at present. OBJECTIVE:To summarize the mechanism,application and research progress of biomimetic remineralization of early enamel demineralization,and provide ideas for further overcoming the hot issues of biomimetic remineralization. METHODS:English keywords"enamel demineralization,biomimetic remineralization,amelogenin,amorphous calcium phosphate"were used to search PubMed.Chinese keywords"enamel demineralization,biomimetic remineralization,amelogenin"were used to search CNKI.Through screening,72 articles were finally obtained for review. RESULTS AND CONCLUSION:(1)At present,there are drug treatments for enamel demineralization,such as fluoride preparations,laser treatment,resin penetration,remineralization treatment and other treatment methods.Biomimetic remineralization is the most ideal repair method for early enamel demineralization.(2)In the narrow sense,enamel remineralization refers to the mineral re-deposition inside the enamel after early enamel demineralization.In the broad sense,enamel remineralization includes the mineralization deposition on the surface and inside of enamel.(3)Clinical biomimetic remineralization reagents are mainly composed of amelogenin,non amelogenin,amelogenin peptide,casein phosphopeptide-amorphous calcium phosphate complex,etc.The advantages of protein and peptide materials are that they conform to the physiological mechanism and can generate high-strength remineralized materials by inducing orientation.The disadvantages are that the manufacturing process is relatively complex and the cost is high.The remineralization effect of amorphous calcium phosphate complex is good,but it needs to be combined with other materials to play a role.Other calcium phosphate materials are easy to carry and beautiful,but they are easy to cause the formation of dental calculus.(4)Future research should focus on the following aspects:increasing experimental data and clinical results,and clarifying the indications of various methods;explore more biomimetic remineralization methods and find suitable alternative materials;find a reasonable way to combine materials,so that their advantages and disadvantages complement each other.The portability of clinical application can be strengthened to increase the frequency of daily use,so that short-term experimental conclusions can be supported by long-term clinical data.
9.Construction of machine learning classification prediction model for vancomycin blood concentrations based on MIMIC-Ⅳ database
Xiaohui LIN ; Yujia WANG ; Lingling ZHANG ; Shuanglin XU
China Pharmacy 2025;36(19):2448-2453
OBJECTIVE To construct a classification prediction model for vancomycin blood concentration, and to optimize its precision dosing strategies. METHODS Patient records meeting inclusion criteria were extracted from the Medical Information Mart for Intensive Care database. Following data cleaning and preprocessing, a final cohort of 9 902 patient was analyzed. Feature selection was performed through correlation analysis and the Boruta feature selection algorithm. Vancomycin blood concentrations were discretized into three categories based on clinical therapeutic windows: low (<10 μg/mL), intermediate (10-20 μg/mL), and high (≥20 μg/mL). Six machine learning algorithms were employed to construct classification models: tabular prior-data fitted network (TabPFN), logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), K-nearest neighbors (KNN). Model performance was evaluated using 10-fold cross-validation (10-CV), with primary metrics including: accuracy, balanced accuracy, precision macro, recall macro, macro F1, area under the receiver operating characteristic curve (OvR-AUC). Shapley Additive Explanations (SHAP) was adopted to analyze the direction and magnitude of the impact that different features had on the model’s predictive outcomes. RESULTS The results showed that the RF and TabPFN models performed the best (with accuracy of 0.741 4 and 0.737 7, and OvR-AUC of 0.907 0 and 0.895 8, respectively). XGBoost model exhibited moderate performance, while LR, SVM, and KNN models demonstrated relatively poor performance. Confusion matrix heatmap analysis revealed that both RF and TabPFN achieved higher accuracy in predicting high- concentration cases but exhibited slightly lower performance in the low and medium concentration categories. Bootstrap with 10-CV revealed that the RF model demonstrated stable performance across various evaluation metrics (accuracy: 0.741 4; balanced accuracy: 0.740 3; precision macro: 0.732 1; recall macro: 0.736 0; macro F1: 0.736 0; OvR-AUC: 0.907 0), indicating good classification performance and generalization ability. SHAP analysis revealed that creatinine, urea nitrogen, daily cumulative dose and administration frequency of vancomycin, which were key predictors, had a significant impact on the prediction results. CONCLUSIONS RF and TabPFN models demonstrate certain advantages in the classification prediction of vancomycin trough blood concentrations; however, their performance in the low to moderate concentration categories still requires improvement.
10.Nomogram of hepatocellular carcinoma ultrasonic feature regression model for predicting microvascular invasion
Jiahan DONG ; Lin ZHOU ; Xiaohui WANG
Chinese Journal of Medical Imaging Technology 2025;41(10):1682-1686
Objective To explore the value of nomogram of hepatocellular carcinoma(HCC)ultrasonic feature regression model for predicting microvascular invasion(MVI).Methods A total of 400 HCC patients(432 lesions)confirmed by pathology were retrospectively collected and divided into training set and validation set at the ratio of 7∶3.There were 280 cases(302 lesions)in training set,including 160 cases(172 lesions)of MVI(+)and 120 cases(130 lesions)of MVI(—),while 120 cases(130 lesions)in validation set,including 70 cases(76 lesions)of MVI(+)and 50 cases(54 lesions)of MVI(-).Univariate and multivariate logistic regression analyses were performed to analyze ultrasonic manifestations of HCC.The independent predictors of MVI of HCC were screened out,and a regression model and nomogram were established,and the efficacy,calibration and clinical value of the nomogram for predicting MVI of HCC were evaluated.Results The maximum diameter of HCC(OR=2.564),margin regular or not(OR=0.412),internal echo uniform or not(OR=1.875),the presence or absence of capsule(OR=0.305)and Adler blood flow grade(OR=3.502)were all independent predictors of MVI of HCC(all P<0.05).The sensitivity,specificity,accuracy and the area under the curve(AUC)of the nomogram for predicting MVI of HCC in training set was 84.88%,81.05%,83.11%and 0.950,respectively,while in validation set was 78.95%,77.78%,78.46%and 0.910,respectively.Calibration curve and decision curve analysis indicated that this nomogram had good calibration and high clinical net benefit in both training and validation sets.Conclusion Nomogram of ultrasound feature regression model could be used to effectively predict MVI of HCC.


Result Analysis
Print
Save
E-mail