Fasting <span>blood glucose is used as an indicator in the process of predicting diabetes risk. This research aims to, i) create a model for predicting blood glucose level using data mining algorithms, ii) a selection algorithm was used to select a feature from the correlation of the data, and iii) to compare the model's performance with the classical methods. All clinical data ware recorded and compiled in a database by hospital staff from 2014-2019. In our previous research, the blood glucose prediction model had an acceptable accuracy where 18 patient features were used as input data to the data mining process. In this research, we demonstrated that the random forest classifier and extra tree classifier algorithms have an outstanding in discarding non-critical attributes. And the process of reducing the number of those features has impacted the glycemic prediction model with higher efficiency. Seventeen machine learning algorithms are used to find the best performance models. Our results clearly show that the improved prediction model is more efficient. This experiment has shown that improvements to our proposed model were able to predict blood glucose levels with 99.69% and 99.63% accuracy for random forest classifier, extra tree classifier, and Gaussian process classifier, respectively.</span>

No abstract

No abstract

In this paper, we propose effective method for texture segmentation using active contour model with edge flow vector. This technique was applied from previous active contour model that uses gradient vector flow as external force. It was observed that our method provided better results for texture segmentation while a traditional active contour model and active contour model with gradient vector flow were not capable to be used with texture image. Thus, texture image such as medical imaging can be identified using active contour model with edge flow vector.

Background Floods cause not only damage but also public health issues. Developing an application to simulate public health problems during floods around the Loei River by implementing geographic information system (GIS) and structural equation model (SEM) techniques could help improve preparedness and aid plans in response to such problems in general and at the subdistrict level. As a result, the effects of public health problems would be physically and mentally less severe. Methods This research and development study examines cross-sectional survey data. Data on demographics, flood severity, preparedness, help, and public health problems during floods were collected using a five-part questionnaire. Calculated from the population proportion living within 300 m of the Loei River, the sample size was 560 people. The participants in each subdistrict were recruited proportionally in line with the course of the Loei River. Compared to the empirical data, the data analysis examined the causal model of public health problems during floods, flood severity, preparedness, and help. The standardized factor loadings obtained from the SEM analysis were substituted as the loadings in the equations for simulating public health problems during floods. Results The results revealed that the causal model of public health problems during floods, flood severity, preparation, and help agreed with the empirical data. Flood severity, preparedness, and aid (χ2 = 479.757, df = 160, p value <.05, CFI = 0.985, RMSEA = 0.060, χ2/df = 2.998) could explain 7.7% of public health problems. The computed values were applied in a GIS environment to simulate public health problem situations at the province, district, and subdistrict levels. Conclusions Flood severity and public health problems during floods were positively correlated; in contrast, preparedness and help showed an inverse relationship with public health problems. A total of 7.7% of the variance in public health problems during floods could be predicted. The analysed data were assigned in the GIS environment in the developed application to simulate public health problem situations during floods.