1.Identification and infection rate of densovirus in Culex pipiens pallens in Beijing in 2023
Xiu-yan XU ; Ting YAN ; Si-jie ZHU ; Jing LI ; Mei-de LIU ; Hong-jiang ZHANG ; Ting LIU ; Qiu-hong LI ; Xiao-jie ZHOU ; Ying TONG ; Yong ZHANG
Acta Parasitologica et Medica Entomologica Sinica 2026;33(1):25-30
Objective This study conducted molecular biological identification of the viruses carried by Culex pipiens pallens specimens collected in Shunyi District, Beijing in 2023, and observed the changes in the infection rate of the viruses carried by Cx. pipiens pallens at different collection times. Methods Cx. pipiens pallens were collected using carbon dioxide mosquito traps. The mosquito samples were ground in batches and analyzed by molecular biology technologies. The virus infection rate at different collection times was analyzed statistically. Results 17 strains of Culex pipiens pallens densovirus(CppDNV)were identified from Cx. pipiens pallens samples collected in Shunyi District, Beijing, in 2023. The nucleotide sequence analysis of the virus genome coding region showed that CppDNV was a single-stranded DNA virus with a total length of 3 335 nt, encoding 2 non-structural proteins(NS1, NS2)and 1 capsid protein(VP). The nucleotide(amino acid)sequence lengths of the three proteins were 2 376 nt(791 aa),1 092 nt(363 aa)and 1 071 nt(356 aa), respectively. Phylogenetic analysis showed that CppDNV was located in genus Brevihamaparvovirus. Statistical analysis showed no significant difference in infection rates across collection times(χ2=4.429, P=0.194). Conclusions CppDNV was identified in Cx. pipiens pallens in Beijing, and it was stably maintained in this natural population.
2.Investigation of mosquito species and breeding sources of larvae in the rice planting area of Beijing
Ting YAN ; Xiu-yan XU ; Jing LI ; Xiao-jie ZHOU ; Yong ZHANG ; Ying TONG
Acta Parasitologica et Medica Entomologica Sinica 2026;33(1):49-52
Objective To investigate the species composition of mosquitoes and breeding conditions of larvae in the rice planting areas of Beijing, and to provide a scientific basis for mosquito control in rice fields. Methods Adult mosquitoes were collected using CO2-baited light traps, and mosquito larvae(including pupae)were sampled via the dipping method. Results A total of 4 genera and 6 species of adult mosquitoes were captured in the study area. The dominant species were Culex pipiens pallens, Cx. tritaeniorhynchus, and Anopheles sinensis. The average adult mosquito density was 5.11 specimens per trap-hour, with slightly lower density in July compared to August. For larvae and pupae,3 genera and 4 species were identified, dominated by Cx. tritaeniorhynchus, An. sinensis, and Aedes vexans. Conclusions The mosquito species composition in Beijing′s rice planting areas is diverse. Mosquito control strategies should prioritize the management of adult mosquitoes, while density control of mosquito larvae in paddy fields can be achieved through intermittent irrigation techniques.
3.Frailty status and its influencing factors among middle-aged and elderly individuals of Zhiguo ethnic groups in Yunnan province
Xiaoju LI ; Rong LI ; Xiu XIA ; Yuan HUANG ; Chaofang YAN ; Rui DENG ; Ying SONG
Chinese Journal of Health Management 2025;19(9):721-727
Objective:To analyze the frailty status and its influencing factors among middle-aged and elderly individuals of Zhiguo ethnic groups in Yunnan Province.Methods:A cross-sectional study was conducted using a multi-stage random cluster sampling method in the settlements of the Zhiguo ethnic groups in Yunnan Province from July to December in 2022. The study enrolled a total of 1 036 middle-aged and elderly individuals aged 45 years and above. Data on general characteristics (including chronic disease prevalence, sleep quality, smoking and alcohol consumption) and frailty status were collected in the subjects. A total of 1 050 questionnaires were distributed, 1 040 questionnaires returned with 1 036 valid responses (98.7%). Logistic regression analysis was performed to identify factors influencing frailty status among different Zhiguo ethnic groups.Results:Among the 1 036 middle-aged and elderly individuals of Zhiguo ethnic groups in Yunnan Province, there were 482 males and 554 females, with a frailty rate of 26.25% (272/1 036), the Bulang ethnic group had the highest frailty rate (35.56%) and the Jinuo ethnic group showed the lowest (18.49%). Being aged≥60 years ( OR=1.647, 95% CI: 1.177-2.303), having≥2 chronic diseases ( OR=2.183, 95% CI: 1.470-3.245), and suffering from sleep disorders ( OR=2.468, 95% CI: 1.786-3.411) were positively associated with frailty occurrence, whereas having high social support ( OR=0.614, 95% CI: 0.434-0.867) was inversely associated with the risk of frailty (all P<0.05). Ethnic subgroup analyses revealed that suffering from sleep disorders were a common influencing factor for frailty across all Zhiguo ethnic groups (all P<0.05). In the Jinuo ethnic group, having≥2 chronic diseases showed a positive association with frailty ( OR=3.387, 95% CI: 1.054-10.879) ( P<0.05). In the Bulang subgroup, having≥2 chronic diseases ( OR=3.951, 95% CI: 1.201-12.995) and aged ≥60 years ( OR=3.415, 95% CI: 1.347-8.657) were both positively associated with frailty, while having high social support ( OR=0.185, 95% CI: 0.065-0.527) was inversely associated with the risk of frailty (all P<0.05). Conclusion:The frailty rate in middle-aged and elderly individuals among Zhiguo ethnic groups in Yunnan Province is notably high, and sleep quality, social support, and chronic conditions are its main influencing factors.
4.Association of urine cadmium levels with thyroid hormone levels among middle-aged and older adults aged 40-89 years in selected areas of China
Changzi WU ; Xiaochen WANG ; Yue CHEN ; Zheng LI ; Yi ZHANG ; Yuan WEI ; Bing WU ; Wenli ZHANG ; Zhengxiong YANG ; Xiaojie DONG ; Ruiting HAO ; Xiu YE ; Luxi WEI ; Yingli QU ; Haiyan CHU ; Yuebin LYU ; Ying ZHU ; Dongqun XU ; Xiaoming SHI
Chinese Journal of Preventive Medicine 2025;59(2):209-215
Objective:To explore the relationship between urinary cadmium levels and thyroid hormone levels in people aged 40-89 years old in selected areas of China.Methods:Based on the "Investigation of the Impact of Soil Quality of Agricultural Land on Human Health in Typical Areas" project from October 2019 to August 2020, a multi-stage stratified random sampling method was used to include 6 588 middle-aged and older adults aged 40-89. Demographic characteristics, dietary frequency and disease status were collected through the questionnaire and physical examination. Urinary cadmium and urinary creatinine were detected by random midstream urine. Fasting venous blood was collected for the detection of Triiodothyronine (T3) and Thyroxine (T4). The linear mixed effects model was used to explore the association of urine cadmium levels with thyroid hormone levels. Its dose-response relationship was explored by using the restricted cubic spline.Results:The age of the subjects was (63.48±12.18) years, with males accounting for 51.28%. The M ( Q 1,Q 3) of urinary cadmium level, T3 and T4 was 2.48 (1.36, 4.42) μg/g·creatinine, (1.96±0.51) nmol/L and (113.75±29.11) nmol/L, respectively. The linear mixed effects model showed that the changes of T3 and T4 were 0.027 (0.009, 0.044) nmol/L and 2.019 (1.084, 2.953) nmol/L for each one-unit increase (natural logarithm transformed) of urinary cadmium. The restricted cubic spline showed that there was a positive nonlinear association between urinary cadmium and T3 as well as T4 (all Pnonlinear<0.05). Conclusion:In selected areas of China, the urinary cadmium level of middle-aged and older adults aged 40-89 years is positively associated with T3 and T4.
5.Analysis of the intervention effects of skill training for parents with autism child on toddlers with autism spectrum disorder
Qinyi YE ; Bingrui ZHOU ; Ying ZHANG ; Chunchun HU ; Yingzhi GU ; Xueyan LI ; Huiping LI ; Qiong XU ; Feiyong JIA ; Lan ZHANG ; Xiu XU
Chinese Journal of Pediatrics 2025;63(5):491-497
Objective:To explore the intervention effects of the skill training for parents with autism child (STPAC) on toddlers with autism spectrum disorder (ASD).Methods:A multicenter non-randomized concurrent controlled study design was conducted. Thirty children with ASD aged 15-30 months, first diagnosed at the Children′s Hospital of Fudan University, the First Hospital of Jilin University, and Chengdu Women′s and Children′s Central Hospital from 2019 to 2020, were enrolled in the STPAC group. Thirty children with ASD who visited the same hospitals during the same period but refused the STPAC intervention were selected as the control group. The STPAC group received an 8-week intervention (3 h/week) followed by quarterly follow-ups for 1 year, while the control group voluntarily chose community-based routine interventions. The Griffiths development scales-Chinese (GDS-C) was used to assess the developmental levels, and the communication and symbolic behavior scales developmental profile infant-toddler checklist (CSBS-DP-ITC) was completed by the primary caregivers to evaluate social, language and symbolic behavior. The independent samples t-tests or Mann-Whitney U tests, etc.was used for inter-group comparison. The paired t-tests or Wilcoxon signed-rank tests, etc. was used for inter-group pre-post intervention comparison. Results:The STPAC group included 30 children (22 males and 8 females, aged (23.9±2.2) months), and the control group included 30 children (20 males and 10 females, aged (24.2±2.6) months). Before the intervention, there were no statistically differences in GDS-C development quotient (DQ) and CSBS-DP-ITC scores between groups (all P>0.05). After 1-year intervention, GDS-C DQ in personal-social, hearing-language, hand-eye coordination, performance domains of STPAC group and GDS-C DQ in personal-social, hearing-language domains of control group were all increased (all P<0.01). After 1-year intervention, CSBS-DP-ITC scores of both groups were all improved in socia, language, symbolic behavior, and total scores (all P<0.001). GDS-C DQ changes before and after 1 year of intervention in hearing-language, hand-eye coordination, performance domains of the STPAC group were all higher the those of control group (34(15, 48 vs. 10(-4, 39), 11±20 vs. -1±19, 23±25 vs. 8±22, all P<0.05). CSBS-DP-ITC scores changes before and after 1 year of intervention in social and total scores of the STPAC group were both higher the those of control group (10(5, 30) vs. 3(1, 7), 26±17 vs. 11±8, both P<0.001). Conclusion:Compared with the community routine interventions, the STPAC better improves the language, hand-eye coordination, visual-spatial, social communication, and play skills in ASD toddlers.
6.Effect of miR-185-5p targeted negative regulation of TM9SF1 on proliferation,migration and autophagy in lung adenocarcinoma cells
Xiao-Na WANG ; Xiu-Ying GONG ; Miao-Miao ZHAO ; Qing-Hua LIU ; Yong LI ; Kun WANG ; Chong-Gao YIN ; Hong-Li LI
Medical Journal of Chinese People's Liberation Army 2025;50(5):566-574
Objective To investigate the effect of miR-185-5p-mediated targeted negative regulation of transmembrane 9 superfamily member 1(TM9SF1)on proliferation,migration and autophagy in lung adenocarcinoma cells.Methods The expression of miR-185-5p in lung adenocarcinoma tissues was analyzed using dataset GSE51853 downloaded from the Gene Expression Omnibus(GEO)database.Potential target proteins of miR-185-5p were predicted using online databases(miRTargetLink,miRTarbase,and DIANA-microT-CD),and autophagy-related proteins were obtained from HADb.The intersected results from these four databases was identified,and survival curves of vascular endothelial growth factor A(VEGFA)and TM9SF1 within the overlapping candidates were analyzed using the StarBase database.TM9SF1 3'UTR wild-type(WT)or TM9SF1 3'UTR mutant(MUT)reporter plasmids were separately co-transfected with miR-185-5p control plasmid(CON)or miR-185-5p overexpression plasmid(over-miR-185-5p)into HEK-293T cells.A dual-luciferase reporter gene assay was employed to assess the binding interaction between miR-185-5p and TM9SF1 and quantify the subsequent luciferase activity.Western blotting was used to assess TM9SF1 protein expression levels in A549 cells transfected with over-miR-185-5p.A549 cells were divided into three groups:(1)CON+NC group,co-transfected with miR-185-5p control plasmid and TM9SF1 control plasmid;(2)over-miR-185-5p+NC group,co-transfected with over-miR-185-5p and TM9SF1 control plasmid;(3)over-miR-185-5p+over-TM9SF1 group,co-transfected with both miR-185-5p and TM9SF1 overexpression plasmids.EdU cell proliferation assay,wound healing assay,and Transwell migration assay were performed to validate the effects of miR-185-5p targeted binding to TM9SF1 on proliferation and migration capacities in lung adenocarcinoma.Changes in autophagic flux and mitochondrial membrane potential(MMP)of lung adenocarcinoma cells were detected using stubRFP-sensGFP-LC3 lentivirus and JC-1 assays,respectively.Results In the GSE51853 dataset,miR-185-5p expression level was significantly lower in lung adenocarcinoma tissues compared with normal lung tissues(P<0.01).qRT-PCR analysis revealed that miR-185-5p expression was downregulated in lung adenocarcinoma cell lines NCI-H1299 and A549 compared with normal lung epithelial cells BEAS-2B(P<0.01).Bioinformatics predictions using miRTargetLink,miRTarbase,DIANA-microT-CD,and HADb databases indicated that miR-185-5p could target and regulate the autophagy-related protein TM9SF1.Dual-luciferase reporter assays and Western blotting demonstrated that miR-185-5p directly bound to the 3'UTR region of TM9SF1 mRNA,and overexpression of miR-185-5p significantly reduced the expression of target protein TM9SF1(P<0.05).EdU cell proliferation,wound healing,and Transwell migration assays demonstrated that miR-185-5p overexpression inhibited proliferation and migration capacities of lung adenocarcinoma cells,whereas TM9SF1 overexpression could attenuate this inhibition effect(P<0.05).Results of stubRFP-sensGFP-LC3 for autophagic flux analysis demonstrated that overexpression of miR-185-5p enhanced autophagic flux in A549 cells,whereas co-overexpression of miR-185-5p and TM9SF1 suppressed autophagic flux.JC-1 assays showed a decreased MMP level in A549 cells after miR-185-5p overexpression,with higher MMP level observed when miR-185-5p and TM9SF1 were co-overexpressed.Conclusion miR-185-5p may suppress proliferation,migration,and autophagy capacities in lung adenocarcinoma cells by targeting TM9SF1 through negative regulation.
7.Clinical and pathological characteristics of adrenal cortical carcinoma:a single-center retrospective study
Qing-Zheng WU ; Ming-Xiu YANG ; Bing LI ; Shu-Ying LI ; Zi-Xin GUO ; Yi-Jun LI ; Ya-Qi YIN ; Ya-Jing WANG ; Kang CHEN ; Li ZANG ; Wei-Jun GU ; Yi-Ming MU ; Zhao-Hui LYU
Medical Journal of Chinese People's Liberation Army 2025;50(7):786-792
Objective To investigate the clinical and pathological characteristics of adrenal cortical carcinoma(ACC),compare differences between hypercortisolism and non-functional ACC,and assess the diagnostic value of indicators such as Ki-67 index.Methods The clinical data of 57 ACC patients admitted to the First Medical Center of Chinese PLA General Hospital from January 2015 to March 2025 were retrospectively analyzed.According to the results of endocrine function assessment,47 of these patients were divided into hypercortisolism group(n=19)and non-functional group(n=28).The differences in clinical and pathological characteristics between the two groups were compared,and non-parametric tests and Spearman correlation analysis were used to explore the relationship between Ki-67 index and tumor stage as well as imaging features.Results Among the 57 patients,there were 20 males and 37 females,with a male-to-female ratio of 1:1.85.The age ranged from 16 to 76 years,and the age at diagnosis was(48.7±13.3)years.The tumor diameter was(10.53±4.14)cm.The tumors were located on the right side in 12 cases(21.1%),on the left side in 34 cases(59.6%),and bilaterally in 11 cases(19.3%).Among them,16 cases(28.1%)were complicated with glucose metabolism disorders,31 cases(54.3%)had hypertension,and 20 cases(35.1%)had hypokalemia.According to ENSAT staging,there were 0 cases in stage Ⅰ,15 cases(26.3%)in stage Ⅱ,24 cases(42.1%)in stage Ⅲ,and 18 cases(31.6%)in stage Ⅳ.Endocrine function assessment was completed in 47 of the 57 patients,including 28 cases(59.6%)of non-functional ACC and 19 cases(40.4%)of hypercortisolism(including 1 case of hypercortisolism combined with increased sex hormone secretion).Compared with non-functional group,hypercortisolism group had a significantly higher prevalence of hypertension(P=0.014),later ENSAT stage(P=0.010),and a higher proportion of hypervascularization(P=0.048).The median Ki-67 index was 20%(10%-40%),showing no significant correlation with either the maximum tumor diameter or SUVmax value,but it was related to ENSAT staging,with Ki-67 index in stageⅣ patients being significantly higher than that in stage Ⅱ(P=0.032).Immunohistochemistry results showed that the positive rate of Inhibin-α was 84.8%,and the positive rate of Melan-A was 40.9%.Conclusions ACC is a rare malignant endocrine tumor.ACC patients with hypercortisolism are more likely to be complicated with hypertension,have later staging,and more common hypervascular manifestations.Clinically,their endocrine function should be prioritized for assessment,and more active treatment strategies should be adopted.Diagnosis should be combined with imaging characteristics(such as hypervascularization)and immunohistochemical indicators(Ki-67,Inhibin-α,Melan-A).The significant increase in Ki-67 is in the advanced stage can serve as an important prognostic indicator to guide individualized treatment.
8.The efficacy of plasma gasdermin D C-terminal fragment in the early diagnosis of sepsis
Yuexian LYU ; Xiu BI ; Ying LIU ; Shujing CUI ; Lixin ZHAO ; Ge GAO ; Jianxia WANG ; Juan LI ; Jun LI
The Journal of Practical Medicine 2025;41(12):1899-1906
Objective To assess the effectiveness of the Gasdermin D C-terminal fragment(GSDMD-CT)as a novel plasma biomarker for the clinical diagnosis of sepsis.Methods Between July 2021 and November 2024,245 patients from Tangshan Gongren Hospital were enrolled in this study.In accordance with the diagnostic criteria for sepsis and the systemic inflammatory response syndrome(SIRS),patient samples were classified into the sepsis group and the SIRS group.Meanwhile,healthy individuals were selected as the healthy control(HC)group.Sepsis patients were further categorized into the Gram-positive bacterial group,the Gram-negative bacterial group,and the fungal group based on the type of pathogen infection.The levels of GSDMD-CT,C-reactive protein(CRP),and procalcitonin(PCT)were measured in all subjects.Nonparametric tests were employed to compare the differences in various indices among different groups.The diagnostic value of GSDMD-CT in sepsis was evalu-ated by constructing the receiver operating characteristic(ROC)curve.Spearman's correlation analysis was used to examine the relationships among GSDMD-CT,CRP,and PCT.Results The plasma GSDMD-CT levels in the sepsis group 23.02(16.71,33.01)pg/mL and in the SIRS group 16.52(11.26,22.22)pg/mL were significantly higher than those in the healthy control group 7.02(4.42,11.43)pg/mL(U=-10.175,-7.890,P<0.001).Moreover,the plasma GSDMD-CT levels in the sepsis group were significantly higher than those in the SIRS group(U=-2.941,P<0.05).In the Gram-positive bacterial group,the Gram-negative bacterial group,and the fungal group,the GSDMD-CT levels were 23.01(17.16,27.51)pg/mL,23.41(16.78,35.50)pg/mL,and 16.29(14.53,56.27)pg/mL,respectively.When compared with the healthy control group,the GSDMD-CT levels in these three groups were all significantly higher(P<0.05).However,there were no significant differences in GSDMD-CT levels among these three groups(P>0.05).The area under the curve(AUC)of plasma GSDMD-CT for diagnosing sepsis was 0.881(95%confidence interval:0.833~0.929),with a Youden index(YI)of 0.695,a sensitivity of 85.0%,and a specificity of 84.5%.Spearman correlation analysis indicated a weak correlation between GSDMD-CT and C-reactive protein(CRP)(r=0.32,P<0.001)and a positive correlation between GSDMD-CT and procalci-tonin(PCT)(r=0.65,P<0.001).Conclusion GSDMD-CT exhibits significant clinical value in the diagnosis of sepsis and holds great potential as a biomarker in the diagnostic process of sepsis.
9.Characterization of the genetic evolution of tick-borne spotted fever group rickettsiae in selected areas of Qinghai
Zhi LI ; Hai-ning ZHANG ; Xue-yong ZHANG ; Hong DUO ; Xiu-ying SHEN ; Hong YIN ; Yong FU ; Zhi-hong GUO
Chinese Journal of Zoonoses 2025;41(4):419-426
The study was aimed at identifying the diversity of tick species in selected areas of Qinghai,to analyze the genetic differentiation characteristics of tick-borne spotted fever group rickettsiae(SFGR),and to provide the theoretical basis for SFGR prevention and control in the region.The 16S rRNA gene was used for molecular biological identification of 446 collected tick samples,and the infection characteristics of SFGR in tick samples were determined according to the SFGR outer membrane protein A(ompA)gene.Haplotype analysis,phylogenetic tree construction,and estimation of differentiation times for SFGR were conducted in DNASP v6,IQ-tree v2.2.0,and BEAST v2.7.4 software.The obtained 446 tick samples belonged to three categories:(1)Haemaphy-salis spp.,including Haemaphysalis qinghaiensis(n=192)and H.danieli(n=37);(2)Dermacentor spp.,including Dermacentor ever-estianus(n=121),D.nuttalli(n=55),and D.silvarum(n=36);and(3)Hyalomma marginatum(n=5).Rickettsia raoultii was de-tected in D.everestianus,D.silvarum,D.nuttalli,H.qinghaiensis,and H.danieli,with infection rates of 95.9%,80.6%,69.1%,4.1%,and 2.7%,respectively.R.sibirica subsp.sibirica BJ-90 was found only in D.silvarum and D.nuttalli,with infection rates of 5.6%and 1.8%,respectively.The Candidatus R.gannanii F107 was found in H.danieli and H.qinghaiensis,with infection rates of 16.2%and 7.8%,respectively.Ca.R.hongyuanensis was detected only in H.qinghaiensis,with a prevalence of 16.3%.The prevalence of R.aeschlimannii was 20%and 2.7%in Hy.marginatum and H.danieli,respectively.Haplotype and nucleotide polymorphism analy-ses revealed 13 haplotypes in R.raoultii,with haplotype H13 as the dominant haplotype(42/192);seven haplotypes in Ca.R.ganna-nii F107,with haplotype H4 as the dominant haplotype(4/18);and three haplotypes in Ca.R.hongyuanensis,with haplotype H1 as the dominant haplotype(11/13).The phylogenetic tree indicated that the sequences of R.raoultii in selected areas of Qinghai and R.rhipicephali clustered into one branch;Ca.R.hongyuanensis and Ca.R.gannanii F107 clustered into one branch;and R.sibirica subsp.sibirica BJ-90 clustered into one branch with R.sibirica.Estimates of differentiation time revealed that the mean differentiation time for the six Rickettsia was approximately 2 000 Mya(95%CI:1 999.08-2 001.02 Mya).The tick species distributed in selected ar-eas of Qinghai are diverse,and this study provides the first report of Hy.marginatum in Qinghai Province.SFGR significantly varied in prevalence among tick species and showed high genetic diversity.
10.Construction and Optimization of Alzheimer's Disease Classification Model Based on Brain Mixed Function Network Topology Parameters and Machine Learning
Xiao-yu HAN ; Xiu-zhu JIA ; Yang LI ; Meng-ying LOU ; Yong-qi NIE ; Xin-ping GUO ; Lu YU ; Zhi-yuan LI ; Lian-zheng SU
Progress in Modern Biomedicine 2025;25(11):1770-1778
Objective:To explore the interrelationship between brain functional networks and features in functional magnetic resonance imaging(fMRI)of patients with Alzheimer's disease(AD),and to construct mixed-function networks(MFN),and apply them in machine learning classification models to improve the accuracy of AD classification.Methods:102 AD patients and 227 healthy subjects in the Alzheimer's Neuroimaging Initiative(ADNI)dataset were retrospectively analyzed.The partial correlation brain network of the blood oxygen level dependent(BOLD)signal was calculated and fused with low-frequency wave amplitude(ALFF),fractional low-frequency wave amplitude(fALFF)and local consistency(ReHo)features to construct MFN.Network topology parameters were extracted,and a variety of machine learning classification models were constructed based on MFN topological parameters,accuracy,precision,recall and area under the curve(AUC)were used to evaluate the predictive efficiency of the models.Results:By constructed MFN and calculated intra group to inter group ratio(IIGR),35 features could be obtained from ALFF,fALFF and ReHo feature topological parameter analysis,after rank sum test and FDR correction,there were statistical differences among 28 features(P<0.05).The classification results show that,all the five classifiers have high classification performance on the test data set.The accuracy,precision and recall rates of random forest(RF),adaptive lifting algorithm(AdaBoost),guided aggregation algorithm(Bagging)and support vector machine(SVM)were all 99.7%,and the AUC values were up to 100%,99.5%,99.1%and 99.5%,respectively.The accuracy(98.5%),precision(98.5%),recall(98.5%),and AUC(99.1%)of the multi-layer perceptron(MLP)were slightly lower than other models,but remained excellent.It was worth noting that RF has the highest AUC value of all models at 100.0%,while Bagging has the lowest AUC value(99.1%)in the integrated approach.The results of performance comparison show that,MFN classification model can significantly improve the recognition and classification of AD disease,and greatly improve the performance of various indicators of the classifier.The results showed that,MFN classification model was superior to intelligent classification based fusion,DBN-based multitask learning,PVT-TSVM,unsupervised learning and clustering,SVM and SVM of degree 3 polynomial kernel function in key indicators such as accuracy(99.13%),AUC(99.42%),recall rate(99.46%)and specificity(99.42%)with plasma proteins,machine learning algorithms.It was further proved that MFN classification model has good generalization ability and robustness in AD disease classification.Conclusion:The AD classification model constructed based on brain mixed function network topology parameters and machine learning can improve the accuracy of AD classification.


Result Analysis
Print
Save
E-mail