Scholarlink

The classification of DNA sequences remains a significant challenge for biologists seeking to analyze vast amounts of biological data. Artificial intelligence has emerged as a powerful tool for addressing this issue, revolutionizing healthcare in various ways. From developing new clinical systems to managing patient data and treating diseases, artificial intelligence techniques, such as machine learning and deep learning, are essential. The COVID-19 pandemic has highlighted the critical need for accurate and efficient disease identification. Machine learning approaches have proven invaluable in predicting COVID-19 infections. In this paper, we employed complete human gene sequences to differentiate COVID-19 (SARS-CoV-2) from other coronavirus strains. We utilized Long-Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), and Autoencoder techniques for classification. Five models were proposed: CNN, LSTM, CNN-LSTM, AE-CNN, and AE-CNN-LSTM. Our hybrid AE-CNN-LSTM model demonstrated superior performance, achieving an overall accuracy of 99% in COVID-19 classification.

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 Hybrid Deep Learning Model Using CNN, LSTM and Auto Encoder to Predict Covid-19 from Human Genome

  • 章节

  • 发表日期：

    2025年

查看全文
  • Noha El-Attar,
  • Bossy M. Moustafa,
  • Sami A. AbdelHafeez,
  • Wael A. Awad

摘要

The classification of DNA sequences remains a significant challenge for biologists seeking to analyze vast amounts of biological data. Artificial intelligence has emerged as a powerful tool for addressing this issue, revolutionizing healthcare in various ways. From developing new clinical systems to managing patient data and treating diseases, artificial intelligence techniques, such as machine learning and deep learning, are essential. The COVID-19 pandemic has highlighted the critical need for accurate and efficient disease identification. Machine learning approaches have proven invaluable in predicting COVID-19 infections. In this paper, we employed complete human gene sequences to differentiate COVID-19 (SARS-CoV-2) from other coronavirus strains. We utilized Long-Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), and Autoencoder techniques for classification. Five models were proposed: CNN, LSTM, CNN-LSTM, AE-CNN, and AE-CNN-LSTM. Our hybrid AE-CNN-LSTM model demonstrated superior performance, achieving an overall accuracy of 99% in COVID-19 classification.