KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Access type: UnknownSubject: search.filters.subject.Bacteria

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Deep Learning to Classify Bacterial Species in the same Genus
    (2024-01-01)
    Sheela, Sherin
    ;
    Piang, May Phu
    ;
    Sakorntanant, Sakda
    ;
    Poomrittigul, Suvit
    ;
    Treebupachatsakul, Treesukon
    Bacterial strains in the same genus share highly similar morphology, gram-staining characteristics, colony sizes, and spatial arrangements. Therefore, identifying them by deep learning can be quite challenging. This study aimed to assess the classification of 7 species of bacteria from 2 genera of Bacillus and Vibrio by using 8 Convolutional Neural Network (CNN) models. We implemented Python programming along with Keras API within the Jupyter Notebook. The models were constructed and evaluated under unbalanced and balanced datasets by augmentation (rotation, flip, etc.). Transfer learning with fine-tuning, and pre-processing of mixup and label smoothing were also applied to reduce overfitting and enhance generalization. Based on the experimental results on private dataset, the results of InceptionResNetV2 emerged as the top-performing model with a notable accuracy of 82.8%, 88.6% precision, 78.4% recall, and 78.0% F1-score when label smoothing was applied at 0.5 on balanced dataset.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A Comparison of Deep Learning CNN Architecture Models for Classifying Bacteria
    (2022-01-01)
    Poomrittigul, Suvit
    ;
    Chomkwah, Wanwalee
    ;
    Tanpatanan, Tananan
    ;
    Sakorntanant, Sakda
    ;
    Treebupachatsakul, Treesukon
    Since identifying bacteria from a patient's sample for medical diagnosis purposes by the traditional approach is time-consuming and requires the pathologist's expertise to do the bacteria identification procedure. Thus, involving the deep learning model reported the capability of multi-class image classification allows us to reduce the time and increase the prediction accuracy of the bacteria identification process. This research includes 35 different bacteria species and 6 different Convolutional Neural Network (CNN) architectures. Convolutional Neural Network (CNN) architectures are LeNet-5, AlexNet, VGG-16, VGG-19, ResNet-18, and ResNet-34. The results confirmed the perceptional performance by applying Stratified K-fold cross validation with VGG-16 and observing the multi-class performance with the AUC-ROC score.