Scholarlink

Amidst the backdrop of the COVID-19 pandemic and the pressing global health challenge of lung disease, this study presents a pioneering approach to bolster diagnostic evaluations. Focused on harnessing the power of deep learning techniques, our research endeavors to navigate the intricacies of lung disease diagnosis. A novel hybrid CNN model is introduced, tailored to proficiently identify and scrutinize lung abnormalities, with specific attention to the unique complexities posed by COVID-19. Through comparative analyses against standalone CNN models like VGG16, ResNet50, VGG19and InceptionV3, we showcase the superiority of our hybrid model. Remarkably, our proposed model attains heightened efficiency while employing fewer parameters and effectively addressing overfitting concerns commonly encountered in deep learning methodologies. This study represents a significant stride forward in the realm of lung disease diagnostics, furnishing a more robust and precise approach for comprehensive assessment, particularly within the COVID-19 landscape. This advancement contributes substantially to the evolution of sophisticated diagnostic tools, ultimately enhancing patient care and outcomes in the global battle against lung disease.

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

Improving Pneumonia Diagnosis via Deep Learning: A Comprehensive Approach Incorporating CNN Classification

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • S. Kumar Reddy Mallidi,
  • RamaDevi Vanapalli,
  • D. S. Priyanka Jella,
  • Tejaswi Nimmakayala,
  • Devi Narni,
  • Chandini Attili

摘要

Amidst the backdrop of the COVID-19 pandemic and the pressing global health challenge of lung disease, this study presents a pioneering approach to bolster diagnostic evaluations. Focused on harnessing the power of deep learning techniques, our research endeavors to navigate the intricacies of lung disease diagnosis. A novel hybrid CNN model is introduced, tailored to proficiently identify and scrutinize lung abnormalities, with specific attention to the unique complexities posed by COVID-19. Through comparative analyses against standalone CNN models like VGG16, ResNet50, VGG19and InceptionV3, we showcase the superiority of our hybrid model. Remarkably, our proposed model attains heightened efficiency while employing fewer parameters and effectively addressing overfitting concerns commonly encountered in deep learning methodologies. This study represents a significant stride forward in the realm of lung disease diagnostics, furnishing a more robust and precise approach for comprehensive assessment, particularly within the COVID-19 landscape. This advancement contributes substantially to the evolution of sophisticated diagnostic tools, ultimately enhancing patient care and outcomes in the global battle against lung disease.