Scholarlink

<p>In recent years, the incidence of lung diseases has increased significantly, leading to millions of cases annually. These diseases, including conditions such as asthma, emphysema, pulmonary embolism, COVID-19, non-COVID pneumonia, and lung cancer, are often dangerous and potentially fatal. Many of these diseases share similar symptoms, such as difficulty breathing and chest tightness, making accurate diagnosis challenging. To address this, we propose a novel multi-class lung disease detection model based on a Depth-wise Separable Convolutional Neural Network (DSCNN) and a Multilayer Perceptron (MLP) for detecting COVID-19, non-COVID pneumonia, and lung cancer from CT images. Our approach includes preprocessing, feature extraction, and classification. Initially, images from the dataset are fed into the preprocessing stage, where Gaussian Filtering (GF) is applied to reduce noise. The DSCNN model is then used for feature extraction, followed by the MLP for classifying different lung diseases. We evaluated the proposed method using publicly available CT image datasets across four classifications: normal, COVID-19, pneumonia, and lung cancer. The implementation was done using Python 3, and the model’s performance was assessed based on accuracy, sensitivity, specificity, precision, recall, and F1 score.</p>

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

A Deep Learning-Based Multiple Lung Disease Prediction and Classification Using CT- Scan Images

  • 期刊论文

  • 发表日期：

    2025年4月9日

查看全文
  • V. Helen Deva Priya,
  • A. Vimala Juliet

摘要

In recent years, the incidence of lung diseases has increased significantly, leading to millions of cases annually. These diseases, including conditions such as asthma, emphysema, pulmonary embolism, COVID-19, non-COVID pneumonia, and lung cancer, are often dangerous and potentially fatal. Many of these diseases share similar symptoms, such as difficulty breathing and chest tightness, making accurate diagnosis challenging. To address this, we propose a novel multi-class lung disease detection model based on a Depth-wise Separable Convolutional Neural Network (DSCNN) and a Multilayer Perceptron (MLP) for detecting COVID-19, non-COVID pneumonia, and lung cancer from CT images. Our approach includes preprocessing, feature extraction, and classification. Initially, images from the dataset are fed into the preprocessing stage, where Gaussian Filtering (GF) is applied to reduce noise. The DSCNN model is then used for feature extraction, followed by the MLP for classifying different lung diseases. We evaluated the proposed method using publicly available CT image datasets across four classifications: normal, COVID-19, pneumonia, and lung cancer. The implementation was done using Python 3, and the model’s performance was assessed based on accuracy, sensitivity, specificity, precision, recall, and F1 score.