Scholarlink

Cardiomegaly detection from chest X-rays is an important process in diagnosing heart disease, but it requires expertise from physicians. This research aims to improve the efficiency of Cardiomegaly Detection using deep learning technology with image processing. It presents the use of Contrast Limited Adaptive Histogram Equalization (CLAHE), Color Histogram, and grayscale to binary image conversion by finding the optimal threshold value (Otsu Thresholding) together with the use of the ResNet-50 deep learning model to detect cardiac enlargement from chest X-rays. The results of the research show that the process of combining image processing methods with the use of the ResNet-50 model can detect cardiac enlargement accurately and is highly efficient in classifying abnormal cardiac characteristics. In addition, the automation system helps reduce the diagnosis time and increases the consistency of the evaluation results, which is very useful to assist physicians and increase the quality of treatment. This research demonstrates the potential of applying modern technology to detect cardiac enlargement from X-ray.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cardiac Hypertrophy Detection Using ResNet-50: A Deep Learning Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sokliv Kork,
  • Katanyu Biadklang,
  • Parkpoom Chaisiriprasert

摘要

Cardiomegaly detection from chest X-rays is an important process in diagnosing heart disease, but it requires expertise from physicians. This research aims to improve the efficiency of Cardiomegaly Detection using deep learning technology with image processing. It presents the use of Contrast Limited Adaptive Histogram Equalization (CLAHE), Color Histogram, and grayscale to binary image conversion by finding the optimal threshold value (Otsu Thresholding) together with the use of the ResNet-50 deep learning model to detect cardiac enlargement from chest X-rays. The results of the research show that the process of combining image processing methods with the use of the ResNet-50 model can detect cardiac enlargement accurately and is highly efficient in classifying abnormal cardiac characteristics. In addition, the automation system helps reduce the diagnosis time and increases the consistency of the evaluation results, which is very useful to assist physicians and increase the quality of treatment. This research demonstrates the potential of applying modern technology to detect cardiac enlargement from X-ray.