Comparison of Different Machine Learning Techniques to Predict Diabetic Kidney Disease.

Background: Diabetic kidney disease (DKD), one of the complications of diabetes in patients, leads to progressive loss of kidney function. Timely intervention is known to improve outcomes. Therefore, screening patients to identify high-risk populations is important. Machine learning classification techniques can be applied to patient datasets to identify high-risk patients by building a predictive model. Objective: This study aims to identify a suitable classification technique for predicting DKD by applying different classification techniques to a DKD dataset and comparing their performance using WEKA machine learning software. Methods: The performance of nine different classification techniques was analyzed on a DKD dataset with 410 instances and 18 attributes. Data preprocessing was carried out using the PartitionMembershipFilter. A 10-fold cross validation was performed on the dataset. The performance was assessed on the basis of the execution time, accuracy, correctly and incorrectly classified instances, kappa statistics (K), mean absolute error, root mean squared error, and true values of the confusion matrix. Results: With an accuracy of 93.6585% and a higher K value (0.8731), IBK and random tree classification techniques were found to be the best performing techniques. Moreover, they also exhibited the lowest root mean squared error rate (0.2496). There were 15 false-positive instances and 11 false-negative instances with these prediction models. Conclusions: This study identified IBK and random tree classification techniques as the best performing classifiers and accurate prediction methods for DKD.

Effect of a very low-carbohydrate ketogenic diet vs recommended diets in patients with type 2 diabetes: a meta-analysis.

CONTEXT: There is renewed interest in using very low-carbohydrate ketogenic (VLCK) diets to manage diabetes. Many clinical trials have been published, often with mixed results. OBJECTIVE: This meta-analysis compares the effect of a VLCK diet on glycemic control, body weight, lipid profile, medication use, and dropouts with that of recommended diets for 12 weeks or longer in people with type 2 diabetes. DATA SOURCES: Ovid MEDLINE, Ovid Embase, CENTRAL, and CINAHL databases were searched (January 1980 through September 2019). STUDY SELECTION: Two authors independently reviewed search results to select randomized controlled trials (RCTs) comparing a VLCK diet (carbohydrate intake < 50 g/d or < 10% of total energy) with any recommended diet for type 2 diabetes in adults. Discrepancies were resolved after consulting with the third author. DATA EXTRACTION: Eight RCTs with 648 participants were identified. RESULTS: Compared with control diets, the VLCK diet resulted in a greater decrease in hemoglobin A1c after 3 months (weighted mean difference[WMD]: -6.7 mmol/mol; 95%CI, -9.0 to -4.4) (WMD: -0.61%; 95%CI, -0.82 to -0.40; P < 0.001; moderate-certainty evidence) and after 6 months (WMD: -6.3 mmol/mol; 95%CI, -9.3 to -3.5) (WMD: -0.58%; 95%CI, -0.85 to -0.32; low-certainty evidence). There was a significantly greater weight loss with the VLCK diet after 3 months (WMD: -2.91 kg; 95%CI, -4.88 to -0.95; low-certainty evidence) and after 6 months (WMD: -2.84 kg; 95%CI, -5.29 to -0.39; low-certainty evidence). The VLCK diet was not better than a control diet after 12 months. It was superior in decreasing triglyceride levels, increasing high-density lipoprotein cholesterol levels, and reducing the use of antidiabetic medications for up to 12 months. CONCLUSION: The VLCK diet appears to control glycemia and decrease body weight for up to 6 months in people with obesity and diabetes. Beneficial changes in serum triglycerides and high-density lipoprotein cholesterol, along with reductions in antidiabetic medications, continued in the VLCK group until 12 months. However, the quality of currently available evidence is not sufficient to recommend VLCK diets. A major limitation of the VLCK diet is patients' lack of adherence to carbohydrate restriction. SYSTEMATIC REVIEW REGISTRATION: PROSPERO registration number CRD42020154700.

Health apps usage and preferences among Saudi patients with diabetes: A survey.

BACKGROUND: Information regarding patterns of smartphone usage, user preferences, and difficulties faced by patients is necessary to initiate any mHealth related programs. OBJECTIVE: To investigate patterns of usage pertaining to various different smartphone health applications amongst diabetic patients in Saudi Arabia as well as patient preferences and the challenges of using health apps from a user perspective. METHODS: This study was a cross sectional survey based on a self-administered, close-ended, pre-designed, structured questionnaire. The main emphasis of the questions were as follows: (a) to investigate current patterns of smartphone use, (b) to examine the use of mobile for health applications and (c) to understand patient preferences when it comes to the specific health related use of mobile phone apps (d) to accrue patient feedback in using health apps. RESULTS: 36.62% of the study population reported using health applications. Patients preferred to follow up blood sugar measurements, body weight, exercise accomplishments, and caloric intake through a health application. More than a third of the participants who were surveyed found the health apps to be very easy to understand with an equal number of participants responding that some training was either necessary. Approximately 32.54% of the respondents indicated that they did not want to know about their health. However, almost an equal number of participants responded that they were unsure of how to start on health apps. CONCLUSION: The results of this survey indicate that there is huge potential for utilizing these health applications via smartphones in the implementation of various health programs.

Correction to: "In Silico Detection of Virulence Gene Homologues in the Human Pathogen Sphingomonas Spp.".

[This corrects the article on p. 229 in vol. 10, PMID: 25574122.].

Comparative Estimation of Genetic Diversity in Population Studies using Molecular Sampling and Traditional Sampling Methods.

Entomopathogenic nematodes (EPN) are efficient biological pest control agents. Population genetics studies on EPN are seldom known. Therefore, it is of interest to evaluate the significance of molecular sampling method (MSM) for accuracy, time needed, and cost effectiveness over traditional sampling method (TSM). The study was conducted at the Mohican Hills golf course at the state of Ohio where the EPN H. bacteriophora has been monitored for 18 years. The nematode population occupies an area of approximately 3700 m(2) with density range from 0.25-2 per gram soil. Genetic diversity of EPN was studied by molecular sampling method (MSM) and traditional sampling method (TSM) using the mitochondrial gene pcox1. The MSM picked 88% in compared to TSM with only 30% of sequenced cox 1 gene. All studied genetic polymorphism measures (sequence and haplotype) showed high levels of genetic diversity of MSM over TSM. MSM minimizes the chance of mitochondrial genes amplification from non target organisms (insect or other contaminating microorganisms). Moreover, it allows the sampling of more individuals with a reliable and credible representative sample size. Thus, we show that MSM supersedes TSM in labour intensity, time consumption and requirement of no special experience and efficiency.

In silico detection of virulence gene homologues in the human pathogen sphingomonas spp.

There is an ongoing debate about the clinical significance of Sphingomonas paucimobilis as a virulent bacterial pathogen. In the present study, we investigated the presence of different virulence factors and genes in Sphingomonas bacteria. We utilized phylogenetic, comparative genomics and bioinformatics analysis to investigate the potentiality of Sphingomonas bacteria as virulent pathogenic bacteria. The 16S ribosomal RNA gene (16S rDNA) phylogenetic tree showed that the closest bacterial taxon to Sphingomonas is Brucella with a bootstrap value of 87 followed by Helicobacter, Campylobacter, Pseudomonas, and then Legionella. Sphingomonas shared no virulence factors with Helicobacter or Campylobacter, despite their close phylogenic relationship. In spite of the phylogenetic divergence between Sphingomonas and Pseudomonas, they shared many major virulence factors, such as adherence, antiphagocytosis, iron uptake, proteases, and quorum sensing. In conclusion, Sphingomonas spp. contains several major virulence factors resembling Pseudomonas sp., Legionella sp., Brucella sp., and Bordetella sp. virulence factors. Similarity of virulence factors did not match phylogenetic relationships. These findings suggest horizontal gene transfer of virulence factors rather than sharing a common pathogenic ancestor. Sphingomonas spp. is potential virulent bacterial pathogen.