Clinical Characteristics and Detection Sensitivity of Cervical Cancer Screening in Vaginal Intraepithelial Neoplasia.

OBJECTIVE: This study aimed to investigate the characteristics and screening history of vaginal intraepithelial neoplasia (VaIN) or vaginal cancer and compare the sensitivity of cytology and human papillomavirus (HPV) tests on the cervix against vaginal and cervical high-grade squamous intraepithelial lesion or cancer. METHODS: This study included patients who underwent colposcopy-directed biopsy and were diagnosed with VaIN or vaginal cancer from February 2013 to November 2022. Clinical information was obtained from the medical records of the department. Statistical analysis was performed on SPSS 26.0 (IBM Corp, Armonk, NY) using t test, chi-square, and Fisher exact tests. RESULTS: A total of 1,166 patients were included in this study. The median age of VaIN2+ patients was 50.5 years, whereas VaIN1 reported a median age of 42.1 years old, p < .001. This study reported that VaIN was significantly and positively correlated with cervical lesions (r = 0.244). The high-risk HPV (hr-HPV) detection rate was 88.2% (858/973) in VaIN and 95.2% in VaIN2+. Human papillomavirus 16 was the most prevalent HPV type in VaIN2+, which accounted for 54.9%, followed by HPV58 (19.5%), HPV52 (15.2%), HPV51 (12.2%), and HPV18 (11.0%). The sensitivity of hr-HPV and cytology tests on the cervix for detecting VaIN2+ was 94.7% and 83.4%, respectively. Both tests were not significantly different from detecting cervical intraepithelial neoplasia 2+. CONCLUSIONS: Human papillomavirus 16 is the dominant HPV type in vaginal precancer lesions. Cervical cancer screening has similar sensitivity for VaIN2+ as for cervical intraepithelial neoplasia 2+, with hr-HPV testing showing higher sensitivity than cytology.

Prevalence of multiple human papillomavirus infections and association with cervical lesions among outpatients in Fujian, China: A cross-sectional study.

Multiple human papillomavirus (HPV) infections are common, but their impact on cervical lesions remains controversial. A total of 6225 female patients who underwent colposcopies/conization following abnormal cervical cancer screening results were included in the study. The final pathological diagnosis was determined by the most severe pathological grade among the cervical biopsy, endocervical curettage, and conization. Univariate and multivariate logistic regression analyses were used to investigate the association between multiple HPV infections and cervical lesions, adjusting for age, HPV genotype, gravidity and parity. In total, 33.3% (n = 2076) of the study population was infected with multiple HPV genotypes. Multiple HPV infections were more prevalent in patients younger than 25 years and older than 55 years, with the rate of multiple HPV infections at 52.8% and 44.3%, respectively. HPV16\52\18\58 are the most common HPV genotypes and usually appear as a single infection. Compared to single HR-HPV infection, multiple HR-HPV infections do not increase the risk of HSIL+, while single HR-HPV coinfected with LR-HPV seems to reduce the risk of HSIL+ (odds ratio = 0.515, confidence interval: 0.370-0.719, p < 0.001). Multiple HR-HPV infections cannot be risk-stratified for triage of HR-HPV-positive women.

Prevotella as the hub of the cervicovaginal microbiota affects the occurrence of persistent human papillomavirus infection and cervical lesions in women of childbearing age via host NF-κB/C-myc.

There is evidence that coinfection of cervicovaginal high-risk human papillomavirus (HR-HPV) and bacteria is common in women of childbearing age. However, the relationship between bacterial vaginosis (BV) and persistent HR-HPV infection in women of childbearing age and the underlying mechanisms remain unclear. In this study, we determined whether BV affects persistent HR-HPV infection in women aged 20-45 years and explored the possible mechanisms of their interactions. From January 1 to April 30, 2020, we recruited women aged 20-45 years with and without BV at a ratio of 1:2 from Fujian Maternity and Child Health Hospital. All women were followed up at 0, 12, and 24 months. A BV assay, HR-HPV genotyping and cervical cytology were performed at each follow-up. At 0 months, additional vaginal secretions and cervical exfoliated cells were collected for 16S ribosomal RNA sequencing, bacterial metabolite determination, and POU5F1B, C-myc, TLR4, NF-κB, and hTERT quantification. A total of 920 women were included. The abundance of Prevotella (p = 0.016) and Gardnerella (p = 0.027) were higher, whereas the abundance of Lactobacillus was lower (p = 0.001) in women with persistent HR-HPV infection and high-grade squamous intraepithelial lesions (HSIL). The abundance of Prevotella (p = 0.025) and Gardnerella (p = 0.018) increased in the vaginas of women with persistent HPV16 infection, whereas only the abundance of Prevotella (p = 0.026) was increased in women with persistent HPV18 infection. The abundance of Prevotella in the vagina was significantly positively correlated with the expression levels of TLR4, NF-κB, C-myc, and hTERT in host cervical cells (p < 0.05). Our findings suggest that overgrowth of Prevotella in the vagina may influence the occurrence of persistent HR-HPV infection-related cervical lesions through host NF-κB and C-myc signaling.

Cervical Lesion Classification Method Based on Cross-Validation Decision Fusion Method of Vision Transformer and DenseNet.

Objective: In order to better adapt to clinical applications, this paper proposes a cross-validation decision-making fusion method of Vision Transformer and DenseNet161. Methods: The dataset is the most critical acetic acid image for clinical diagnosis, and the SR areas are processed by a specific method. Then, the Vision Transformer and DenseNet161 models are trained by the fivefold cross-validation method, and the fivefold prediction results corresponding to the two models are fused by different weights. Finally, the five fused results are averaged to obtain the category with the highest probability. Results: The results show that the fusion method in this paper reaches an accuracy rate of 68% for the four classifications of cervical lesions. Conclusions: It is more suitable for clinical environments, effectively reducing the missed detection rate and ensuring the life and health of patients.

Changes of the vaginal microbiota in HPV infection and cervical intraepithelial neoplasia: a cross-sectional analysis.

This study aimed to explore the changes of the vaginal microbiota and enzymes in the women with high-risk human papillomavirus (HR-HPV) infection and cervical lesions. A total of 448 participants were carried out HPV genotyping, cytology tests, and microecology tests, and 28 participants were treated as sub-samples, in which vaginal samples were characterized by sequencing the bacterial 16S V4 ribosomal RNA (rRNA) gene region. The study found the prevalence of HR-HPV was higher in patients with BV (P = 0.036). The HR-HPV infection rate was 72.73% in G. vaginalis women, which was significantly higher than that of women with lactobacillus as the dominant microbiota (44.72%) (P = 0.04). The positive rate of sialidase (SNA) was higher in women with HR-HPV infection (P = 0.004) and women diagnosed with cervical intraepithelial neoplasia (CIN) (P = 0.041). In HPV (+) women, the α-diversity was significantly higher than that in HPV (-) women. The 16S rRNA gene-based amplicon sequencing results showed that Lactobacillus was the dominant bacteria in the normal vaginal microbiota. However, the proportion of Gardnerella and Prevotella were markedly increased in HPV (+) patients. Gardnerella and Prevotella are the most high-risk combination for the development of HPV (+) women. The SNA secreted by Gardnerella and Prevotella may play a significant role in HPV infection progress to cervical lesions.

Stabilization of Antitumor Agent Busulfan through Encapsulation within a Water-Soluble Pillar[5]arene.

The complexation of the antitumor agent busulfan by negatively charged carboxylatopillar[5]arene in water is reported. The encapsulation within carboxylatopillar[5]arene reduces the hydrolytic degradation of busulfan from 90.7% to 25.2% after 24 days and accordingly enhances its stability by providing a hydrophobic shelter for busulfan in water. Moreover, the complexation results in 12 times improvement of water solubility for busulfan. Our result provides a supramolecular approach for stabilizing the anticancer agent busulfan.

Integration of Global and Local Features for Specular Reflection Inpainting in Colposcopic Images.

Objective: To explore an inpainting method that can balance texture details and visual observability to eliminate the specular reflection (SR) regions in the colposcopic image, thus improving the accuracy of clinical diagnosis for cervical cancer. Methods: (1) To ensure smoothness, Gaussian Blur and filling methods are applied to the global image. (2) Striving to preserve the anatomical texture details of the colposcopic image as much as possible, the exemplar-based method is applied to local blocks. (3) The colposcopic images inpainted in the previous two steps are integrated, so that important information of non-SR regions is preserved based on eliminating SR regions. Results: In the subjective visual assessment of inpainting results, the average of 3.55 ranks first in the five comparison sets. As to the clinical test, comparing the diagnosis results of 6 physicians before and after eliminating SR regions, the average accuracy of two kinds of classifications increased by 1.44% and 2.03%, respectively. Conclusions: This method can effectively eliminate the SR regions in the colposcopy image and present a satisfactory visual effect. Significance. As a preprocessing method for computer-aided diagnosis systems, it can also improve physicians' accuracy in clinical diagnosis.

Prevalence of Human Papillomavirus and Genotype Distribution in Pregnant and Non-Pregnant Women in China.

PURPOSE: The status of human papillomavirus (HPV) infection in pregnant and non-pregnant women in China remains unclear. This study aimed to compare the prevalence and genotype distributions of HPV between pregnant and non-pregnant women in China. PATIENTS AND METHODS: A case-control study was conducted of pregnant women during the second trimester and age-matched non-pregnant women attending the Fujian Maternity and Child Health Hospital between January 1, 2017 and December 31, 2018. Participants underwent cervical cytology testing and HPV genotyping. The genotyping test was able to identify 14 high-risk HPV (HR-HPV), four possible HR-HPV, and five low-risk HPV (LR-HPV) types. Further colposcopy and a cervical biopsy were performed if indicated. The primary outcomes were HPV prevalence and genotype distribution. RESULTS: In total, 1077 pregnant and 1077 non-pregnant women were enrolled. Compared with non-pregnant women, pregnant women had a higher prevalence of HPV (24.2% vs 14.8%), HR-HPV (20.2% vs 11.7%), and LR-HPV (8% vs 4.5%) infection. In pregnant women, the most prevalent HPV genotypes were HPV-52 (6.0%), -16 (3.5%), -58 (2.6%), -53 (2.5%), and -51 (2.5%), while in non-pregnant women the most prevalent genotypes were HPV-52 (3.6%), -81 (1.9%), -51 (1.8%), -68 (1.4%), and -16 (1.3%). In women aged ≥35 years, HR-HPV (P=0.002) and LR-HPV (P=0.001) prevalence were significantly higher in pregnant women. However, in women aged <35 years, only HR-HPV prevalence was higher in pregnant women. Pregnant and non-pregnant women with HPV-16 and HPV-58 infection had a high prevalence of high-grade squamous intra-epithelial lesions (HSIL) (HPV-16: P<0.001 and P=0.005, HPV-58: P=0.043 and P=0.005); but with other HR-HPV genotypes, only non-pregnant women had an increased HSIL prevalence. CONCLUSION: In China, the HPV prevalence is higher in pregnant women than that in non-pregnant women and is also age- and genotype-dependent. HPV-infected pregnant women aged ≥35 years and those with HPV-16 should be closely monitored to enable rapid clinical intervention.

Use of extended HR-HPV Genotyping in improving the Triage Strategy of 2019 ASCCP recommendations in Women with positive HR-HPV diagnosis and Simultaneous LSIL Cytology Results.

Objective: According to the 2019 American Society for Colposcopy and Cervical Pathology (ASCCP) recommendations, women with a positive high-risk human papillomavirus (HR-HPV) diagnosis and low-grade cervical intraepithelial lesion (LSIL) cytology result should be referred for further colposcopy examination. However, this strategy results in over-treatment in several cases. In this study, we assessed the performance of extended HR-HPV genotyping in women with a simultaneous positive HR-HPV and LSIL diagnosis with the aim of improving the current triage strategy. Methods: This study was an observational analysis of women from the Fujian Province Cervical Lesion Screening Cohorts (FCLSCs). Women who were HR-HPV-positive and had a cytological examination of LSIL, which were followed up with colposcopy and biopsy, from 2015 to 2018 were included. The study endpoint was defined as the detection of histological cervical intraepithelial neoplasia grade 2 or worse (CIN2+). We combined HR-HPV genotypes according to the prevalence rate in histological CIN2+ and ranked them from high to low to establish HR-HPV genotyping models. Outcomes were assessed with respect to sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and colposcopy referral rate. Results: Overall, 56,788 women undergoing preliminary screening for HR-HPV genotyping were included in this study. Among them, 10,499 women positive for HR-HPV underwent a cytology examination, and 902 women with LSIL cytology diagnosed and subsequent biopsy results were included in the final evaluation. Among these patients, 25.1% (226/902) were found to have CIN2+ in histology. HPV-16, -58, -52, -18, -33, and -31 infections were the most common genotypes, and HPV-16, -18, -58, -33, and -31 (odds ratio [OR] = 5.41, 2.98, 1.38, 1.24, and 1.21, respectively) were associated with the potential for histological CIN2+, from the highest to lowest. In the detection of CIN2+ lesions in HR-HPV-positive LSIL women of different HR-HPV genotyping models, the extended HPV 16/18/31/33/52/58 genotyping model was found to have better efficacy with higher sensitivity (92.9%) and NPV (93.0%), but a significantly lower colposcopy referral rate (74.7%) than the ASCCP-recommended HR-HPV non-genotyping model. Conclusion: For HR-HPV-positive women with LSIL, the HPV 16/18/31/33/52/58 genotyping model can serve as an alternative approach to the ASCCP recommendations, potentially reducing the unnecessary colposcopy referral burden in China.

Effect of introducing human papillomavirus genotyping into real-world screening on cervical cancer screening in China: a retrospective population-based cohort study.

BACKGROUND: China's Fujian Cervical Pilot Project (FCPP) transitioned cervical cancer screening from high-risk human papillomavirus (HR-HPV) nongenotyping to genotyping. We investigated the clinical impact of this introduction, comparing performance indicators between HR-HPV genotyping combined with cytology screening (HR-HPV genotyping period) and the previous HR-HPV nongenotyping combined with cytology screening (HR-HPV nongenotyping period). METHODS: A retrospective population-based cohort study was performed using data from the FCPP for China. We obtained data for the HR-HPV nongenotyping period from 1 January 2012 to 31 December 2013, and for the HR-HPV genotyping period from 1 January 2014 to 31 December 2016. Propensity score matching was used to match women from the two periods. Multivariable Cox regression was used to assess factors associated with cervical intraepithelial neoplasia of grade 2 or worse (CIN2+). The primary outcome was the incidence of CIN2+ in women aged ⩾25 years. Performance was assessed and included consistency, reach, effectiveness, adoption, implementation and cost. RESULTS: Compared with HR-HPV nongenotyping period, in the HR-HPV genotyping period, more CIN2+ cases were identified at the initial screening (3.06% versus 2.32%; p < 0.001); the rate of colposcopy referral was higher (10.87% versus 6.64%; p < 0.001); and the hazard ratio of CIN2+ diagnosis was 1.64 (95% confidence interval, 1.43-1.88; p < 0.001) after controlling for health insurance status and age. The total costs of the first round of screening (US$66,609 versus US$65,226; p = 0.293) were similar during the two periods. Higher screening coverage (25.95% versus 25.19%; p = 0.007), higher compliance with age recommendations (92.70% versus 91.69%; p = 0.001), lower over-screening (4.92% versus 10.15%; p < 0.001), and reduced unqualified samples (cytology: 1.48% versus 1.73%, p = 0.099; HR-HPV: 0.57% versus 1.34%, p < 0.001) were observed in the HR-HPV genotyping period. CONCLUSIONS: Introduction of an HR-HPV genotyping assay in China could detect more CIN2+ lesions at earlier stages and improve programmatic indicators. Evidence suggests that the introduction of HR-HPV genotyping is likely to accelerate the elimination of cervical cancer in China.

APLN: A potential novel biomarker for cervical cancer.

Apelin (APLN) is recently demonstrated a direct association with many malignant diseases. However, its effects on cervical cancer remain unclear. This study therefore aims to evaluate the association between APLN expression and cervical cancer using publicly available data from The Cancer Genome Atlas (TCGA). The Pearson χ2 test and Fish exact test, as well as logistic regression, were used to evaluate the relationship between clinicopathological factors in cervical cancer and the expression of APLN. Additionally, the Cox regression and Kaplan-Meier methods were conducted to analyze the Overall Survival (OS) of cervical cancer patients in TCGA. Finally, gene set enrichment analysis (GSEA) was performed to establish its biological functions. High expression of APLN in cervical cancer was significantly associated with a more advanced clinical stage (OR = 1.91 (1.21-3.05) for Stage II, Stage III, and Stage IV vs Stage I, p = 0.006). Additionally, it was associated with poor outcome after primary therapy (OR = 2.14 (1.03-4.59) for Progressive Disease (PD), Stable Disease (SD), and Partial Response (PR) vs Complete Remission (CR), p = 0.045) and high histologic grade (OR = 1.67 (1.03-2.72) for G3 and G4 vs G1 and G2, p = 0.037). Moreover, multivariate analysis showed that high expression of APLN was associated with a shorter OS. GSEA demonstrated that six KEGG pathways, including PPAR signaling, ECM-receptor interaction, focal adhesion, MAPK signaling, TGF-beta signaling, and Gap junction pathways were differentially enriched in the high expression APLN phenotype. The recent study suggests that APLN plays an important role in the progression of cervical cancer and might be a promising prognostic biomarker of the disease.

Integrated Decision-Making Method for Heterogeneous Attributes Based on Probabilistic Linguistic Cross-Entropy and Priority Relations.

The meta-synthesis method has achieved good results in China's aerospace engineering and population economic regulation. This theoretical achievement obtained from engineering practice becomes an effective way to solve complex decision-making problems. The meta-synthesis method obtains the final decision-making result by comprehensively considering qualitative and quantitative criteria and gathering multivariate heterogeneous attribute information. In view of the broad application of entropy theory in quantitative evaluation and fuzzy decision-making, this paper proposes a meta-synthesis decision-making method based on probabilistic linguistic cross-entropy and priority relations for multicriteria decision-making problems including qualitative and quantitative multivariate heterogeneous attribute information. First, the quantitative attribute weight is calculated based on the entropy weight method, and the qualitative attribute weight is calculated by considering the individual effects and interactions of the probabilistic linguistic term sets under qualitative attributes comprehensively through probabilistic linguistic entropy and cross-entropy. Then, the weight preference coefficient is used to integrate the qualitative and quantitative heterogeneous attribute weights to obtain standardized processing weight information, and, on the basis of the 0-1 priority relation matrix, we compare and analyze the advantages and disadvantages of alternatives under all criteria and obtain an overall ranking result of the alternatives. Finally, the effectiveness and superiority of the proposed method are verified by a comparative analysis of a numerical example and the decision-making method.

HPV-16 E2/E6 and POU5F1B as Biomarkers to Determine Cervical High-Grade Squamous Lesions and More.

PURPOSE: Human papillomavirus-16 (HPV-16) is the most carcinogenic HPV genotype. This study aimed to evaluate the clinical value of POU5F1B and HPV-16-E2/E6 by cervical cytology specimens to predict the cervical intraepithelial neoplasia two grade and more (CIN2+). METHODS: Finally, 248 patients with HPV-16 single infection were enrolled. Using cytology specimen by real-time quantitative PCR (qPCR), POU5F1B mRNA and HPV-16-E2/E6 were detected. The relationship of POU5F1B, HPV-16-E2/E6 and CIN2+ were analyzed, and the optimal cut-off values of POU5F1B and HPV-16-E2/E6 to predict CIN2+ were calculated. RESULTS: The mean HPV-16-E2/E6 decreased significantly with cervical lesions development, especially compared with CIN2+ (p<0.05). And the POU5F1B demonstrated higher expression in CIN2+ than that of normal cervical tissue and CIN1 (p<0.05). What is more, POU5F1B was negatively correlated with HPV-16-E2/E6. It demonstrated that the area under the receiver operating characteristic curve (AUC) for POU5F1B (0.9058) was higher than that for HPV-16-E2/E6 (0.8983), and the sensitivity and specificity of POU5F1B in the diagnosis of CIN2+ were higher than HPV-E2/E6. Furthermore, it demonstrated that the POU5F1B had the highest odds ratio (OR= 16.84; 95% CI (8.00-35.46)) for the detection of CIN 2+. CONCLUSION: HPV-16-E2/E6≤0.6471 or POU5F1B≥1.0310 in cervical exfoliated cells can be used as a reliable predictor of CIN2+. POU5F1B can be used as a new auxiliary biomarker to determine the HPV infection status and a reliable predictor of CIN2+. The expression of POU5F1B≥1.0310 had the highest OR for the detection of CIN2+.

The Triage Effectiveness of an Extended High-Risk Human Papillomavirus Genotyping Assay for Women with Cytology Showing Atypical Squamous Cells of Undetermined Significance in China.

PURPOSE: Little is known about the performance of extended high-risk human papillomavirus (HR-HPV) genotyping triage of cytology showing atypical squamous cells of undetermined significance (ASC-US). This study aims to evaluate the effectiveness of triage with different HR-HPV genotype models among women with ASC-US. MATERIALS AND METHODS: In this study, all women who underwent cervical cytology and HR-HPV genotyping were enrolled from 2014 to 2017 in China, and those with cytology showing ASC-US were referred for colposcopy and/or biopsy. The endpoint was histological detection of cervical intraepithelial neoplasia grade 2 or worse (CIN2+). The outcome indicators were the sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs) and colposcopy referral rates. RESULTS: In all, 56,788 women were enrolled in this study, and 2658 (4.97%) women were reported to have ASC-US; 10.1% (242/2393) of women with ASC-US were identified as having CIN2+. The HR-HPV infection rate was 95.0% among all women with ASC-US who were identified as CIN2+, and the top five genotypes with prevalence and risk of CIN2+ were HPV16 (OR=26.38), HPV58 (OR=7.04), HPV18 (OR=4.44), HPV33 (OR=3.38), HPV31 (OR=2.97) and HPV52 (OR=2.96). The HPV16/18/31/33/52/58 model achieved higher sensitivity [91.3 (87.8-94.9)], specificity [70.0 (68.1-72.0)], PPV [25.5 (22.4-28.2)] and NPV [98.6 (97.3-98.7)] for the triage of ASC-US patients than the other HR-HPV-type combination models, but the colposcopy referral rate (36.2%) was significantly lower than that of the recommended HR-HPV nongenotyping model (47.6%). CONCLUSION: This study confirms that the specific HR-HPV genotype HPV16/18/31/33/52/58 is an alternative strategy for ASC-US triage and can effectively reduce the high burden of colposcopy referrals in China.

Cost-effectiveness and accuracy of cervical cancer screening with a high-risk HPV genotyping assay vs a nongenotyping assay in China: an observational cohort study.

BACKGROUND: New screening techniques may affect the optimal approaches for the prevention of cervical cancer. We evaluated the cost-effectiveness and accuracy of alternative screening strategies to provide evidence for cervical cancer screening guidelines in China. METHODS: In total, 32,306 women were enrolled. The current screening with Cervista® high-risk human papillomavirus (HR-HPV) nongenotyping and cytology cotesting (Cervista® cotesting) was compared with PCR-reverse dot blot HR-HPV genotyping and cytology cotesting (PCR-RDB cotesting). All eligible participants were divided into Arm 1, in which both HR-HPV assays were performed, and Arms 2 and 3, in which the PCR-RDB HPV or Cervista® HR-HPV assay, respectively, was performed. Outcome indicators included the cases, sensitivity, negative predictive value (NPV), colposcopy referral rate and cost of identifying cervical intraepithelial neoplasia of grade 2/3 or worse (CIN2+/CIN3+). RESULTS: Among the eligible participants, 18.4% were PCR-RDB HR-HPV-positive, while 16.9% were Cervista® HR-HPV-positive, which reflects good agreement (k = 0.73). PCR-RDB cotesting identified more CIN3+ cases than Cervista® cotesting in the first round of screening in Arm 1 (37 vs 32) and Arms 2/3 (252 vs 165). The sensitivity and NPV of PCR-RDB cotesting for identifying CIN3+ in Arm 1 (sensitivity: 94.9% vs 86.5%; NPV: 99.9% vs 99.7%) and Arms 2/3 (sensitivity: 95.1% vs 80.9%; NPV: 99.9% vs 99.6%) were higher than those of Cervista® cotesting, but the cost was similar. CONCLUSIONS: The PCR-RDB HR-HPV genotyping and Cervista® HR-HPV assay results were consistent. PCR-RDB cotesting possesses optimal cost-effectiveness for cervical cancer screening in China, which has the highest number of cases globally but low screening coverage.

Abnormal Data Region Discrimination and Cross-Monitoring Points Historical Correlation Repair of Water Intake Data.

For the problems of abnormal values existing in the water intake monitoring data and centralized uploaded report, the abnormal data region discrimination (ADRD) algorithm and the cross-monitoring points historical correlation repair (CMHCR) method are proposed to discriminate and repair the abnormal data. The characteristics of abnormal data distribution are analyzed, and the ADRD algorithm is proposed. ADRD uses the relationship between 0 values and the abnormal large value, and the ratio of the abnormal large value to the expectation to distinguish the abnormal data region. The correlation between the monitoring data of current detection points and the historical data of different detection points is analyzed. The results show that the data of current monitoring point and the historical data of corresponding point do not fully conform to the maximum correlation. Therefore, the CMHCR method is proposed to repair abnormal data. Experiments based on actual half year water intake data of 2016 and 2017 are performed by using ADRD. The experimental results show that the proposed algorithm and method can correctly distinguish the abnormal data region and repair the abnormal data properly.

A hybrid ensemble method based on double disturbance for classifying microarray data.

Microarray data has small samples and high dimension, and it contains a significant amount of irrelevant and redundant genes. This paper proposes a hybrid ensemble method based on double disturbance to improve classification performance. Firstly, original genes are ranked through reliefF algorithm and part of the genes are selected from the original genes set, and then a new training set is generated from the original training set according to the previously selected genes. Secondly, D bootstrap training subsets are produced from the previously generated training set by bootstrap technology. Thirdly, an attribute reduction method based on neighborhood mutual information with a different radius is used to reduce genes on each bootstrap training subset to produce new training subsets. Each new training subset is applied to train a base classifier. Finally, a part of the base classifiers are selected based on the teaching-learning-based optimization to build an ensemble by weighted voting. Experimental results on six benchmark cancer microarray datasets showed proposed method decreased ensemble size and obtained higher classification performance compared with Bagging, AdaBoost, and Random Forest.