Scholarlink

<p>The world has witnessed the widespread testing and inspection of various community diseases, with continuous efforts made to curb the transmission of life-threatening infections. While pathogenic laboratory testing remains the gold standard for identifying diseases such as COVID-19, it often involves manual processes and may yield false-negative results. In resource-limited settings, where access to timely and accurate diagnostic tools is constrained, there is a pressing need for automated, rapid, and reliable screening methods. This study proposes a deep learning-based diagnostic model leveraging chest X-ray and computed tomography (CT) images to detect COVID-19 cases efficiently. The model integrates image enhancement techniques and transfer learning, along with noise reduction algorithms, to extract meaningful features and improve detection accuracy. Achieving an accuracy of 98.41%, an F1-score of 98.43%, and a precision of 96.92%, the model demonstrates strong potential for early community-level disease prediction. Such approaches can support frontline healthcare systems by offering a scalable solution that complements traditional testing, ensuring faster decision-making and better healthcare delivery in under-resourced environments.</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

Implementation of Noise Removal and Transfer Learning Technique for the Detection of Community Disease

  • 期刊论文

  • 发表日期：

    2026年2月11日

查看全文
  • Amit Prakash Sen,
  • Anup Kumar,
  • Atul Prakash,
  • Taniya Ghosh

摘要

The world has witnessed the widespread testing and inspection of various community diseases, with continuous efforts made to curb the transmission of life-threatening infections. While pathogenic laboratory testing remains the gold standard for identifying diseases such as COVID-19, it often involves manual processes and may yield false-negative results. In resource-limited settings, where access to timely and accurate diagnostic tools is constrained, there is a pressing need for automated, rapid, and reliable screening methods. This study proposes a deep learning-based diagnostic model leveraging chest X-ray and computed tomography (CT) images to detect COVID-19 cases efficiently. The model integrates image enhancement techniques and transfer learning, along with noise reduction algorithms, to extract meaningful features and improve detection accuracy. Achieving an accuracy of 98.41%, an F1-score of 98.43%, and a precision of 96.92%, the model demonstrates strong potential for early community-level disease prediction. Such approaches can support frontline healthcare systems by offering a scalable solution that complements traditional testing, ensuring faster decision-making and better healthcare delivery in under-resourced environments.