Comparison of Predictive Models for the Early Diagnosis of Diabetes / 대한의료정보학회지

OBJECTIVES: This study develops neural network models to improve the prediction of diabetes using clinical and lifestyle characteristics. Prediction models were developed using a combination of approaches and concepts. METHODS: We used memetic algorithms to update weights and to improve prediction accuracy of models. In the first step, the optimum amount for neural network parameters such as momentum rate, transfer function, and error function were obtained through trial and error and based on the results of previous studies. In the second step, optimum parameters were applied to memetic algorithms in order to improve the accuracy of prediction. This preliminary analysis showed that the accuracy of neural networks is 88%. In the third step, the accuracy of neural network models was improved using a memetic algorithm and resulted model was compared with a logistic regression model using a confusion matrix and receiver operating characteristic curve (ROC). RESULTS: The memetic algorithm improved the accuracy from 88.0% to 93.2%. We also found that memetic algorithm had a higher accuracy than the model from the genetic algorithm and a regression model. Among models, the regression model has the least accuracy. For the memetic algorithm model the amount of sensitivity, specificity, positive predictive value, negative predictive value, and ROC are 96.2, 95.3, 93.8, 92.4, and 0.958 respectively. CONCLUSIONS: The results of this study provide a basis to design a Decision Support System for risk management and planning of care for individuals at risk of diabetes.

Comparison of Predictive Models for the Early Diagnosis of Diabetes / 대한의료정보학회지

OBJECTIVES: This study develops neural network models to improve the prediction of diabetes using clinical and lifestyle characteristics. Prediction models were developed using a combination of approaches and concepts. METHODS: We used memetic algorithms to update weights and to improve prediction accuracy of models. In the first step, the optimum amount for neural network parameters such as momentum rate, transfer function, and error function were obtained through trial and error and based on the results of previous studies. In the second step, optimum parameters were applied to memetic algorithms in order to improve the accuracy of prediction. This preliminary analysis showed that the accuracy of neural networks is 88%. In the third step, the accuracy of neural network models was improved using a memetic algorithm and resulted model was compared with a logistic regression model using a confusion matrix and receiver operating characteristic curve (ROC). RESULTS: The memetic algorithm improved the accuracy from 88.0% to 93.2%. We also found that memetic algorithm had a higher accuracy than the model from the genetic algorithm and a regression model. Among models, the regression model has the least accuracy. For the memetic algorithm model the amount of sensitivity, specificity, positive predictive value, negative predictive value, and ROC are 96.2, 95.3, 93.8, 92.4, and 0.958 respectively. CONCLUSIONS: The results of this study provide a basis to design a Decision Support System for risk management and planning of care for individuals at risk of diabetes.

[Isolation of melittin from bee venom and evaluation of its effect on proliferation of gastric cancer cells]

Background: Gastric cancer [GC] is one of the most common cancers worldwide and in Iran. Conventional therapies are surgery and chemotherapy. Current studies are evaluating natural compounds in inhibiting growth of cancer cell. In this study isolated peptide melittin with 26 amino acids from bee venom and its impact on the viability and proliferation of gastric cancer cells was investigated

Methods: At first melittin was purified from honeybee venom using a reversed-phase high performance liquid chromatography [RP- HPLC] and C18 column. In order to investigate whether melittin, a 26 amino acids peptide which is the main components of honeybee venom, inhibits proliferation of human gastric adenocarcinoma cell line [AGS cells], MTT [[3-[4, 5-dimethylthiazol-2-yl]-2, 5- diphenyltetrazolium bromide] assay was performed. Hemolytic assay carried out in order to confirm the biologic activity of the isolated melittin. AGS cells were plated in a 96-well plate and treated with serially diluted concentrations of melittin for 6 and 12 hours. The mortality of the cells was measured via MTT assay at 540 nm

Results: The obtained chromatogram from RP-HPLC showed that melittin comprises 50% of the studied bee venom. SDS-PAGE analysis of melittin fraction confirmed purity of isolated melittin. Hemolytic activity assay indicates that isolated melittin shows a strong hemolytic activity [HD50=0.5]. MTT assay showed that melittin strongly inhibits proliferation of gastric cancer cells at concentrations more than 2microg/ml. This inhibitory effect is dependent to melittin concentration and incubation time

Conclusion: This study provides evidence that melittin inhibits proliferation of the gastric cancer cells. Results showed that isolated melittin from honey bee venom have cytotoxic effect on AGS cell line with a trend of increasing cytotoxicity with increasing concentration and incubation time

High production of IL-18 by dendritic cells induced by sera from patients with primary antibody deficiency

Predominantly antibody deficiencies are a category of primary immunodeficiency diseases, which consist of several rare disorders such as common variable immunodeficiency [CVID] and X-linked agammaglobulinemia [XLA]. We evaluated the effects of CVID and XLA patients' sera as a source of microenviromental factors on maturation and function of monocyte-derived DCs. Blood was collected from 10 CVID and 5 XLA patients before immunoglobulin replacement therapy and also from 8 healthy volunteers in order to obtain necessary sera for this study. Monocyte derived DCs were generated from blood cells obtained from healthy volunteers in the presence of GM-CSF, IL-4 and 10% serum concentrations from cases and controls. Immature DCs were incubated with monocyte conditioned medium [MCM] and TNF-alpha in order to generate mature DCs. Interleukin 18 [IL-18] production by CD40L-activated mature DCs was measured after 24 hours of culture in vitro. IL-18 production by DCs generated in the presence of CVID and XLA patients' sera were 6.75 +/- 2.59 and 7.08 +/- 1.75 ng/ml, respectively, which were significantly higher than normal serum conditioned DCs [3.55 +/- 0.68] ng/ml. These results suggest that the sera of patients with predominantly antibody deficiencies may contain soluble factor[s] that can induce a significant increase in IL-18 production by DCs