Scholarlink

<p>Current research heavily focuses on using deep learning models to classify images from CT scans and X-rays. This study presents a novel approach that leverages Mask R-CNN, a sophisticated CNN technique known for its effectiveness in object detection and segmentation. I apply this approach to Chest CT Images, greatly improving diagnostic capabilities. Our main objective is to detect and segment ground-glass opacities (GGO) using the MosMedData: COVID19 1000 Dataset, which consists of 1110 cases. The results of our study show that Mask R-CNN has achieved a remarkable accuracy rate of 98.25% and a favorable dice index score. This highlights the promise of Mask R-CNN in improving the analysis of ground-glass opacity (GGO). This study advances the field of medical imaging by showcasing the importance of CT scans in diagnosing COVID-19. It highlights the Mask R-CNN method’s ability to improve the detection and outlining of faint lung abnormalities in CT images. These discoveries offer valuable information for healthcare professionals and radiologists. The source code can be found at <a href="https://github.com/debajyoty/COVID-19-Diagnosis-on-Chest-CT-Images.git">https://github.com/debajyoty/COVID-19-Diagnosis-on-Chest-CT-Images.git</a>.</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

Enhanced detection and segmentation of ground-glass opacities in SARS-CoV-2 patients using Mask R-CNN on Chest CT images

  • 期刊论文

  • 发表日期：

    2025年3月13日

查看全文
  • Debajyoty Banik

摘要

Current research heavily focuses on using deep learning models to classify images from CT scans and X-rays. This study presents a novel approach that leverages Mask R-CNN, a sophisticated CNN technique known for its effectiveness in object detection and segmentation. I apply this approach to Chest CT Images, greatly improving diagnostic capabilities. Our main objective is to detect and segment ground-glass opacities (GGO) using the MosMedData: COVID19 1000 Dataset, which consists of 1110 cases. The results of our study show that Mask R-CNN has achieved a remarkable accuracy rate of 98.25% and a favorable dice index score. This highlights the promise of Mask R-CNN in improving the analysis of ground-glass opacity (GGO). This study advances the field of medical imaging by showcasing the importance of CT scans in diagnosing COVID-19. It highlights the Mask R-CNN method’s ability to improve the detection and outlining of faint lung abnormalities in CT images. These discoveries offer valuable information for healthcare professionals and radiologists. The source code can be found at https://github.com/debajyoty/COVID-19-Diagnosis-on-Chest-CT-Images.git.