Leerojanaprapa, Kanogkan

Loading...
Thumbnail Image
Preferred name
Leerojanaprapa, Kanogkan
Alternative Name
Leerojanaprapa, K.
Main Affiliation
Email
kanogkan.le@kmitl.ac.th

Settings

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Applying Bayesian network for noncommunicable diseases risk analysis: Implementing national health examination survey in Thailand
    (2017-07-02)
    Leerojanaprapa, K.
    ;
    Atthirawong, W.
    ;
    Aekplakorn, W.
    ;
    Sirikasemsuk, K.
    We propose using a Bayesian network to capture and understand the dependency risk factors affecting the prevalence of chronic diseases. By applying a Bayesian network model, we can visualize interdependencies between risks and their effects on the Noncommunicable disease (NCD) prevalence. By using a Bayesian network to model the prevalence of diabetes, we can define the top three risks as family history of diabetes, obesity, and age. Furthermore, the risk classification results can help to determine the managing strategy. For the Thai population, problems arising from family history of diabetes and obesity can be met by employing a transfer strategy. Age (especially ages of 35-59) and the risk incurred by low intake of fruits and vegetables should use a reduction or mitigation strategy. Finally, those at risk as a result of their area of residence (in urban areas) and socio-economic factors within the 4<sup>th</sup> quantile and low level of physical activity should apply a retain strategy.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Comparison of Bayesian Networks for Diabetes Prediction
    (2019-01-01)
    Leerojanaprapa, Kanogkan
    ;
    Sirikasemsuk, Kittiwat
    A Bayesian network (BN) can be used to predict the prevalence of diabetes from the cause&#x2013;effect relationship among risk factors. By applying a BN model, we can capture the interdependencies between direct and indirect risks hierarchically. In this study, we propose to investigate and compare the predictive performances of BN models with non-hierarchical (BNNH), and non-hierarchical and reduced variables (BNNHR) structures, hierarchical structure by expert judgment (BNHE), and hierarchical learning structure (BNHL) with type-2 diabetes. ROC curves, AUC, percentage error, and F1 score were applied to compare performances of those classification techniques. The results of the model comparison from both datasets (training and testing) obtained from the Thai National Health Examination Survey IV ensured that BNHE can predict the prevalence of diabetes most effectively with the highest AUC values of 0.7670 and 0.7760 from the training and the testing dataset, respectively.