Scholarlink

This research puts forth a way of finding the diabetic cardiomyopathy and coronary artery plaque by use of advanced image processing and machine learning techniques. Initially getting microscopic images and resizing them; then turn them into grayscale, enhance them using the Wiener filter and CLAHE. Finally segmentation through Watershed Segmentation which is later refined with LoG filter. For example, energy, entropy, homogeneity, contrast, correlation among others are extracted as features and used to train AlexNB-Net model that combines Naïve Bayes with AlexNet. The model has an accuracy rate of 98.8%, it performs better than XGBoost (87%), Random Forests (81%), Neural Networks (87.5%) or ResNet50 (96.83%). This research offer’s a strong tool for early detection as well as treatment planning for heart diseases.

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

AlexNB-Net: A Novel Hybrid Model for Accurate Classification of Coronary Artery Plaque

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Punganuru Swathi,
  • M. Gunasekaran

摘要

This research puts forth a way of finding the diabetic cardiomyopathy and coronary artery plaque by use of advanced image processing and machine learning techniques. Initially getting microscopic images and resizing them; then turn them into grayscale, enhance them using the Wiener filter and CLAHE. Finally segmentation through Watershed Segmentation which is later refined with LoG filter. For example, energy, entropy, homogeneity, contrast, correlation among others are extracted as features and used to train AlexNB-Net model that combines Naïve Bayes with AlexNet. The model has an accuracy rate of 98.8%, it performs better than XGBoost (87%), Random Forests (81%), Neural Networks (87.5%) or ResNet50 (96.83%). This research offer’s a strong tool for early detection as well as treatment planning for heart diseases.