KMITL

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

Browse

Settings

Author: search.filters.author.Wacharawichanant, SiriratSubject: search.filters.subject.Deep Learning

Search Results

Now showing 1 - 6 of 6
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Improving the Sound Classification Accuracy Using CNN-LSTM and MFCC with Audio Augmentation for Diagnosing Respiratory Disease
    (2025-01-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    Audio is vital information data for understanding various situations. A multitude of sound features can be explained by analysis through the audio signals. Numerous classification methods have been developed to study audio classification. This work studies the improvement of audio classification for the diagnosis of respiratory disease through the integration of audio data augmentation and CNN in conjunction with LSTM (CNN-LSTM). Furthermore, this paper focuses on audio data augmentation and feature extraction in the deep learning approach. This study proposed the CNN-LSTM model to diagnose respiratory disease by learning from the different audio datasets. The results reveal that the CNN-LSTM model attained an accuracy of 81.48%, precision of 0.8340, sensitivity of 0.6948, and F1-score of 0.7225. Considering the achieved F1-score, the CNN-LSTM model demonstrates a high level of diagnotic accuracy. Therefore, all evaluation evaluation parameters collectively indicate the robust performance of the proposed disease classification model.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Subcellular Protein Patterns Classification Using Extreme Gradient Boosting with Deep Transfer Learning as Feature Extractor
    (2024-01-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    Proteins are essential structural and functional components of human cells. Understanding and identifying proteins can provide valuable insights into their structure, function and role in human body. Subcellular proteins provide the expression that characterizes the many proteins and their conditions across cell types. This work proposed a classification model for subcellular protein patterns using XGBoost with transfer learning of CNN as the feature extractor. In the model training process, we used ResNet50, VGG16, Xception, and MobileNet as the pre-trained models based on the transfer learning technique to extract different features. The proposed model was used to classify subcellular proteins into 28 patterns. The XGBoost with ResNet50, VGG16, MobileNet, and Xception model achieved an accuracy level of 92.20% 92.77%, 91.63%, and 91.44%, respectively. The XGBoost with ResNet50, VGG16, MobileNet, and Xception model obtained an F1 score of 0.9179, 0.9233, 0.9131, and 0.9152, respectively. Considering the F1 score, All XGBoost with transfer learning of CNN models gave a high score. Therefore, all evaluation parameters clearly demonstrate the high performance of the subcellular protein pattern classification model.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Improving Classification Efficiency Based on Combination of Extreme Gradient Boosting and Deep Transfer Learning
    (2023-01-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    The leading causes of blindness and low vision are ocular disease. Ocular disease such as glaucoma, cataract, diabetic retinopathy, and macular degeneration, which are diseases in which the risk of vision loss. Unfortunately, some ocular diseases have no symptoms until the late stages. Therefore, early-stage diagnosis of ocular disease is the best way to prevent vision loss. This work proposed the classification models for ocular disease classification using XGBoost in combination with deep transfer learning of CNN as the feature extractor. In the model training process, we used the pre-trained model including Xception and ResNet50 based on the transfer learning technique to extract different features. The proposed model was used to classify ocular disease into eight patterns. The XGBoost in combination with the ResNet50 model achieved an accuracy level of 87.82%, precision of 88.15%, sensitivity of 87.82%, and F1 score of 87.82%. The XGBoost in combination with the Xception model acquired an accuracy level of 87.02%, precision of 87.35%, sensitivity of 87.02%, and F1 score of 87.02%. By considering the F1 score, XGBoost in combination with transfer learning of CNN models gave a high score. Therefore, all evaluation parameters clearly indicate the high performance of the ocular disease classification model. The conclusion presents that the proposed method acquires more excellent performance than individual deep learning models.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Stages of Progression Classification of Alzheimer's Disease Using Deep Transfer Learning Models with Over-Sampling
    (2022-01-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    Alzheimer's disease is a chronic neurodegenerative disease that affected patients loss of memory, ability of thinking, reading, and cognitive decline. The Alzheimer's disease divided stages of progression into four general stages on the basis of their symptoms. The early diagnosis helps to slow down the disease and reduce the costs of treatment. Since Alzheimer's disease has four stages of progression, the classification problems are those where a stage must be predicted in the case of an unequal number of instances of each class. This study has proposed the classification the stages of progression of Alzheimer's disease using four transfer learning models such as VGG19, Xception, ResNet50, and MobileNetV2. The proposed models classify Alzheimer's disease into four stage of progression. The models gained an accuracy level of VGG19, Xception, ResNet50, and MobileNetV2 model of 77.73%, 82.46%, 76.28% and 79.29%, respectively. By considering the F1 score, the Xception, VGG19, and ResNet50, and MobileNetV2 models gave the high score of 0.7995, 0.8870, 0.8305, and 0.5993, respectively. Therefore, the VGG19 model is the best model by considering the F1 score that means the VGG19 model is the best model in overall performance. Finally, this study measures the AUC value that indicates the ability to classify between classes. The results show that AUC value of MobileNetV2, Xception, ResNet50, and VGG19 are 0.9290, 0.9539, 0.7937, and 0.8037, respectively. Therefore, the Xception model is the best model that has capable of distinguishing the stages of progression of the Alzheimer's disease.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    An application of convolutional neural network-long short-term memory model for service demand forecasting
    (2019-07-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    The medical services are very important requirement for being healthy human. In order to ensure the availability of resources for the medicine needed, the most hospital makes an service demand estimation by forecasting a number of patients to provide the sufficient medical services. Therefore, the accurately forecast a number of patients would be valuable knowledge for managing. This work proposed the CNN-LSTM model, which was a combination of CNN and LSTM, to forecast the number of patients who used hospital services. The CNN model was used to interpret, and extract the features from the input data. Then, it was provided this information to the LSTM model for interpreting and making a forecast. The CNN-LSTM models were applied to forecast on the two datasets. The results indicated that CNN-LSTM model made reliable forecasting. This work measured the model performnace by calculating RMSE and MAE value. The result showed RMSE and MAE of the models were very low in all experiments. Forecasting the number of patients can help the hospital to estimate the service demand, make a better policy for managing the medical resources on demand, and improve the efficiency of medical services for the future.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    A comparison of efficiency improvement for long short-term memory model using convolutional operations and convolutional neural network
    (2019-07-01)
    Phankokkruad, Manop
    ;
    Wacharawichanant, Sirirat
    This work studied the comparison of LSTM, ConvLSTM and CNN-LSTM model, that was applied for time series forecasting. We created the LSTM, CNN-LSTM, ConvLSTM model and configured the optimal parameters by using hyperparameters optimization techniques. All models were applied to two different datasets for forecasting the number of patients in the future. This work also applied the SeLu and ReLu activation function to avoid the problem of gradient vanishing and improve the self-normalizing. The results indicated that two models had skillful, and made the reliable forecasting in two datasets. This work benchmarked the model performance by calculating MAE, RMSE, and sMAPE, which was acceptable in all case study. The CNN-LSTM model with SeLu activation function gave highest forecasting efficiency for the data contain seasonal variation. LSTM model with SeLu activation function gave highest forecasting efficiency in the case of non-stationary data.