PlumX Metrics
Embed PlumX Metrics

Akciǧer Hastaliklarinin Teşhisi için Hiyerarşik Siniflamaya Dayali Bir Yöntem

Electrical-Electronics and Biomedical Engineering Conference, ELECO 2024 - Proceedings, Page: 1-5
2024
  • 0
    Citations
  • 0
    Usage
  • 0
    Captures
  • 0
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Conference Paper Description

In this study, a model was developed to diagnose healthy individuals and those with COVID-19 or other viral pneumonia respiratory diseases using hierarchical classification. COVID-19 X-ray images were used, and the VGG19 convolutional neural network (CNN) was employed for classification tasks on these images. The main objective is to achieve high accuracy in distinguishing between healthy and sick cases, followed by detailed classification of disease types. The hierarchical model consists of a primary classifier to separate healthy and sick samples, followed by a sub-classifier that differentiates between COVID-19 and other viral pneumonia cases. Minimal preprocessing steps were applied to achieve high accuracy through this multi-layered classification system, yielding 100% accuracy in distinguishing healthy and sick cases, 97.5% accuracy in sub-classification of diseases, and an overall accuracy of 98.94% for the hierarchical model. These results highlight the potential of hierarchical classification in disease diagnosis using medical imaging.

Bibliographic Details

Rumeysa Yuksel; Sibel Cimen; Bulent Bolat

Institute of Electrical and Electronics Engineers (IEEE)

Engineering

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know