KMITL

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

Browse

Settings

Author: search.filters.author.Sermswan, Anawat

Search Results

Now showing 1 - 3 of 3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Utilizing deep learning from mobile phone photos for early detection of horizontal strabismus: a screening approach
    (2026-12-01)
    Chawuthai, Rathachai
    ;
    Sermswan, Anawat
    ;
    Boonnithititikul, Chatree
    ;
    Hokierti, Kiatthida
    ;
    Sermsripong, Wasawat
    To develop and validate an artificial intelligence pipeline for binary screening of horizontal strabismus versus orthotropia using smartphone-acquired facial images and geometric landmark analysis. This two-stage system combines Real-Time Detection Transformer (RT-DETR) to localize nine ocular landmarks per eye across three gaze directions (left, center, right), and supervised machine learning classifiers. A feature set of five biometric ratios was derived from coordinates including the canthi, limbi, and corneal light reflexes. The model was trained on facial images from 150 participants (96 with strabismus and 54 controls). To address class imbalance and improve generalizability, Synthetic Minority Oversampling Technique (SMOTE) and 4-fold cross-validation were applied. RT-DETR achieved an intersection over union of 0.62 and a mean center-point error of 6.52 pixels in landmark localization. The Random Forest classifier achieved an accuracy of 0.95, sensitivity of 0.96, specificity of 0.94, positive predictive value of 0.97, and negative predictive value of 0.92. This study demonstrates the feasibility of combining transformer-based landmark detection with geometric ratios for strabismus screening. The framework shows high performance under controlled conditions. While the use of biometric ratios allows for feature-level inspection, further research is required to establish full clinical interpretability and performance in uncontrolled environments.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Determining the Optimal Parameters for Preventing Backward Falls Using a Human Movement Dataset: An Investigative Study
    (2024-01-01)
    Yoneyama, Keito R.
    ;
    Sermswan, Anawat
    ;
    Jansri, Ukkrit
    ;
    Srisuttee, Ratakorn
    Falls are a major health concern, particularly for the elderly, as they often lead to severe injuries like hip fractures. Fortunately, there are a lot of publicly available datasets that can offer valuable insights regarding the mechanics of falls and activities of daily living. This study aims to determine the optimal prevention threshold for backward falls using available datasets, thereby avoiding any fall-related injuries. The methodology employed in this study involves identifying downward trends of acceleration and angular velocity on the graph. The Pythagorean Theorem is used to calculate the resultant acceleration, while differences are computed to determine the changes in angular velocity. The results from 150 samples shows that the acceleration for the prevention threshold range is between 3.016m/s<sup>2</sup> - 4.308 m/s<sup>2</sup>and the average angular velocity is 0.522 rad/s - 0.746 rad/s which not only prevents a person from falling backward but will also be able to distinguish between activities of daily living. In conclusion, using existing dataset can offer essential knowledge into fall detection and serves as a foundation for determining the optimal fall prevention threshold.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    The Possible Sarcopenia Associated with Independent Walking Older Adults
    (2023-05-01)
    Raksadawan, Natte
    ;
    Sermswan, Anawat
    ;
    Jansri, Ukkrit
    The amount of skeletal muscle mass and mobility function increase during the childhood, peak at young adult, and then decline after middle age. Older adults with low skeletal mass and mobility function are associated with reduced physical performance. The older adult with a low level of mobility function, regardless of the amount of muscle mass, is defined as "possible sarcopenia". This study aims 1) to characterize the current state of the physical and mobility function at the peak levels of these parameters among young adults 2) to characterize their deterioration rates during middle-age and older-age years and older adults; and 3) to study the prevalence of possible sarcopenia among older adults. The cross-sectional study of anthropometric parameters, body compositions, and mobility assessments was conducted from independent walking young, middle, and older age adults in the community. Young adult male had higher levels of physical and mobility function than female. The changes in physical and mobility function during middle age were subtle. Older age male (female) lost skeletal muscle mass at-0.125 (-0.0190) kg/year, grip strength at-0.495 (-0.307) kg per year, and gait speed at-0.007 (-0.013) m/sec per year, respectively. The prevalence of possible sarcopenia in older age independent walking adults, determined by low grip strength, was 19.4%. The prevalence increased with advancing age.