Advanced methods for missing values imputation based on similarity learning.

The real-world data analysis and processing using data mining techniques often are facing observations that contain missing values. The main challenge of mining datasets is the existence of missing values. The missing values in a dataset should be imputed using the imputation method to improve the data mining methods' accuracy and performance. There are existing techniques that use k-nearest neighbors algorithm for imputing the missing values but determining the appropriate k value can be a challenging task. There are other existing imputation techniques that are based on hard clustering algorithms. When records are not well-separated, as in the case of missing data, hard clustering provides a poor description tool in many cases. In general, the imputation depending on similar records is more accurate than the imputation depending on the entire dataset's records. Improving the similarity among records can result in improving the imputation performance. This paper proposes two numerical missing data imputation methods. A hybrid missing data imputation method is initially proposed, called KI, that incorporates k-nearest neighbors and iterative imputation algorithms. The best set of nearest neighbors for each missing record is discovered through the records similarity by using the k-nearest neighbors algorithm (kNN). To improve the similarity, a suitable k value is estimated automatically for the kNN. The iterative imputation method is then used to impute the missing values of the incomplete records by using the global correlation structure among the selected records. An enhanced hybrid missing data imputation method is then proposed, called FCKI, which is an extension of KI. It integrates fuzzy c-means, k-nearest neighbors, and iterative imputation algorithms to impute the missing data in a dataset. The fuzzy c-means algorithm is selected because the records can belong to multiple clusters at the same time. This can lead to further improvement for similarity. FCKI searches a cluster, instead of the whole dataset, to find the best k-nearest neighbors. It applies two levels of similarity to achieve a higher imputation accuracy. The performance of the proposed imputation techniques is assessed by using fifteen datasets with variant missing ratios for three types of missing data; MCAR, MAR, MNAR. These different missing data types are generated in this work. The datasets with different sizes are used in this paper to validate the model. Therefore, proposed imputation techniques are compared with other missing data imputation methods by means of three measures; the root mean square error (RMSE), the normalized root mean square error (NRMSE), and the mean absolute error (MAE). The results show that the proposed methods achieve better imputation accuracy and require significantly less time than other missing data imputation methods.

Caspase-Cleaved Cytokeratin 18 Fragment M30 as a Potential Biomarker of Macrovascular Invasion in Hepatocellular Carcinoma.

BACKGROUND AND AIM: Extremely poor prognosis in hepatocellular carcinoma (HCC) patients with progressing disease was denoted by vascular invasion. Cytokeratin 18 (CK18) has been shown to be overexpressed in hepatocellular carcinoma so it is a valuable tumor marker; however, its role in vascular invasion is still unclear. This study aimed to predict CK18 as a predictive marker for macrovascular malignant invasion. METHODS: The present study was conducted on three groups of patients: group I included 91 HCC patients without macrovascular invasion, group II included 34 HCC patients with radiological evidence of vascular invasion, and group III included 110 control individuals subdivided into IIIA as healthy blood donors and IIIB as post-HCV cirrhotic patients without HCC. RESULTS: ROC curve of M30 fragments of CK18 was constructed for discrimination between HCC with and without macrovascular invasion. Optimum cutoff value was 304.5 ng/mL (AUC = 0.997, P < 0.001), sensitivity (100%) and specificity (98.8%). Regression analysis was conducted for prediction of macrovascular invasion within HCC patients. The following variables: higher levels of AST, M30, bilirubin, and AFP, lower levels of serum albumin, larger tumor size, child B score, and multiple lesions were associated with vascular invasion in univariate analysis. While in multivariate analysis, higher levels of AST and bilirubin and elevated levels of M30 and AFP serum were considered independent predictors for macrovascular invasion in HCC patients. CONCLUSION: The present study suggests that increased M30 fragments of CK18 levels may be useful as a possible marker of early tumor invasiveness.

Th17 and IL-17 as Predictors of Hepatic Inflammation in Patients with Chronic Hepatitis C Virus Infection and Treated With Direct Antiviral Therapy.

Limited data exists on the role of Th17 cells in chronic HCV infected patients, particularly with regard to hepatic inflammation and fibrosis. We aimed to investigate the relationship between circulating and intrahepatic frequency of Th17 cells and IL-17 serum level and degrees of hepatic inflammation and fibrosis in chronic HCV patients, as well as to evaluate the effect of successful anti-viral therapy on these parameters. This nested longitudinal case control study included 30 treatment-naïve chronic HCV patients and 20 healthy individuals as control. All patients were investigated for circulating Th17 cell percentage (flow cytometry) and intrahepatic Th17 cell percentage (immunohistochemistry) and serum IL-17 (ELISA) at baseline and at week 12 after discontinuation of therapy. Circulating and intrahepatic Th17 cell percentage and serum IL17 level were found to be significantly higher in chronic HCV patients when compared with controls, with significant correlation with Metavir activity score. No patients required discontinuation of therapy due to any adverse event allowing for sustained virological response at 12 weeks (SVR12) in 24 patients while the remaining six patients were considered "non-responders". Circulating Th17 cells and serum IL17 levels were significantly decreased after successful Sofosbuvir-Ribavirin therapy (P < 0.0001). The extent of liver inflammation is positively correlated with frequencies of circulating Th17 cells, and their HCV-specific IL17 secretion, and intrahepatic Th17 cells. This data may also provide the basis for the potential use of Th17 as a new marker for disease advancement of chronic hepatitis C.

Relationship between vitamin A deficiency and the thyroid axis in clinically stable patients with liver cirrhosis related to hepatitis C virus.

Vitamin A deficiency (VAD) and altered thyroid function are commonly encountered in patients with liver cirrhosis. The link between vitamin A metabolism and thyroid function has been previously identified. The aim of this study was to explore the association between VAD and the thyroid axis in clinically stable patients with cirrhosis related to hepatitis C virus (HCV). One hundred and twelve patients with clinically stable HCV-related cirrhosis and 56 healthy controls matched for age, sex, and socioeconomic status were recruited for this study. Vitamin A status, liver function, thyroid-stimulating hormone (TSH), free thyroxine (FT4), free triiodothyronine (FT3), reverse triiodothyronine (rT3), anti-thyroid peroxidase antibodies (anti-TPO), and thyroid volume were evaluated. The prevalence of VAD among patients with HCV-related cirrhosis was 62.5% compared with 5.4% among controls (P < 0.001). Patients with HCV-related cirrhosis had significantly higher FT4, FT3, TSH, and thyroid volume than did healthy controls. Of the 112 patients initially recruited, 18 were excluded (patients with subclinical hypothyroidism and/or anti-TPO positive), so a total of 94 patients with HCV-related cirrhosis were divided into 2 groups according to vitamin A status: VAD and normal vitamin A. Patients with VAD had significantly lower vitamin A intake and serum albumin and higher serum bilirubin, FT4, FT3, and TSH than patients with normal vitamin A status. Multiple logistic regression analysis revealed that VAD was associated with Child-Pugh score (ß = 0.11, P = 0.05) and TSH (ß = -1.63, P = 0.02) independently of confounding variables. We conclude that VAD may be linked to central hyperthyroidism in patients with clinically stable HCV-related liver cirrhosis.

Human leukocyte antigen class I alleles can predict response to pegylated interferon/ribavirin therapy in chronic hepatitis C Egyptian patients.

BACKGROUND: Racial differences and broad spectrum response to anti-hepatitis C (anti-HCV) therapy suggest a possible role for host genetic diversity in treatment outcomes. We aim to determine the association and predictive value of certain human leukocyte antigen (HLA) class I alleles with either susceptibility to viral clearance or persistence following pegylated interferon (Peg-IFN) plus ribavirin therapy in chronic hepatitis C (HCV) genotype 4 patients in Egypt.  METHODS: This study included 200 unrelated chronic HCV patients who received Peg-IFN plus ribavirin therapy [112 patients with sustained virological response (SVR) and 88 non-responders (NR)]. Serological testing of HLA class I antigens (HLA-A and HLA-B alleles) were performed by standard complement-dependent microlymphocytotoxicity assay.  RESULTS: The frequency of HLA-A01 was significantly higher in SVR than in NR cases [OR: 0.51; 95% CI: 0.27-0.981; P = 0.042], while the frequency of alleles B38 (P = 0.011), B40 (P < 0.001) and B41 (P < 0.001) was significantly higher in NR cases (OR/95% CI: 7.05/(1.39-18.01), 10.31/3.14-36.1 . On logistic regression analysis, presence of the HLA-A01 allele was associated with SVR  (OR: 0.50; 95% CI: 0.28-0.89; P = 0.02) and HLA-B38 can predict non response to therapy (OR: 7.92; 95% CI: 1.67-37.54; P = 0.009) with an overall accuracy of 60%.Severe fibrosis (OR: 3.035; 95% CI: 1.521-6.091; P = 0.002), high viremia (OR: 2.69; 95% CI: 1.11-6.53; P = 0.005) and steatosis (OR: 2.1; 95% CI: 1.002-3.90; P = 0.041) predicted no response with an overall accuracy of 81.8%.  CONCLUSION: HLA-A01 and HLA-B38 alleles are associated with and may have a role in the outcome of response to Peg-IFN plus ribavirin therapy in Egyptian patients diagnosed with chronic HCV infection. The use of immunologic markers to predict the outcome of treatment may help pharmacogenetic personalization of treatment for HCV infection.

Colorectal schistosomiasis: Is it still endemic in delta Egypt, early in the third millennium?

BACKGROUND: Several governmental efforts have been exerted toward controlling schistosomiasis during the last decades in Egypt. This work was designed to study the prevalence of colorectal schistosomiasis in patients with different gastrointestinal symptoms. MATERIALS AND METHODS: Patients presented to the gastroenterology unit with different gastrointestinal symptoms were endoscopically examined, where three to six tiny biopsies were taken from those with visible, suspected schistosomal lesions for histopathological examination and two additional rectal biopsies were taken from the apparently normal colonic mucosa. Form each patient, at least three stool samples were examined by the formal-ether concentration method for schistosoma ova. RESULTS: Colonic abnormalities were detected in 510 out of 984 patients presented with different gut symptoms. Schistosoma mansoni was detected in 205 patients (180 males, 25 females) with an age range (18-65years). Six patients only had schistosomal polyps and excised successfully by snare polypectomy. The squash technique established the diagnosis of schistosomiasis in all endoscopically normal 118 (50.75%) cases by demonstrating the schistosomiasis ova and their associated histopathological findings showed no or minimal reaction in 96 (46.82%) cases and variable degrees of submucosal granulomata in the remaining cases. Stool examination detected the schistosomiasis ova in 25 (9.83%) patients only of the biopsy-positive cases. CONCLUSIONS: Our data revealed that despite governmental efforts, the prevalence of colorectal schistosomiasis (20.83%) is significant among patients with gut symptoms. Gaps in health care services should be detected and filled appropriately.