Scholarlink

<p>Medical image enhancement plays an essential role in delivering high-quality medical images. Moreover, healthcare industries are transformed into modern healthcare, where all the medical tools are equipped with the latest computing innovations. Since, medical imaging tools are very helpful to detect diseases and asymptomatic treatment, the image enhancement approach has the ability to convert low-resolution images into high-resolution images, and it has significant importance in scientific research and clinical applications. However, some image enhancement mechanisms can oversmooth the image and lose important details about the initial image, which makes the detection and analysis process critical. Therefore, to enhance the MRI brain image an innovative model, namely Enlighten Generative MIRNet (EGMIRNet) is proposed. Initially, the MRI brain image is given as input for pre-processing, which is done based on Medav filter. Then, the pre-processed image is employed in image enhancement process, which is done by EGMIRNet, the combination of Enlighten Generative Adversarial Network (GAN) and Multiple Identities Representation Network (MIRNet). In addition, the developed EGMIRNet is evaluated based on metrics like Degree of Distortion (DD), Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), and Structural Similarity Index Measure (SSIM), and it has obtained the values of 0.056, 48.187&#xa0;dB, 0.004, and 0.957.</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

EGAN-F-MIRNet: Enlighten Generative Adversarial Network-Fusion-MIRNet for Image Enhancement of MRI Brain Image

  • 期刊论文

  • 发表日期：

    2025年8月29日

查看全文
  • R. Kavitha,
  • R. Arunadevi,
  • K. Ishwarya,
  • S. Athinarayanan,
  • M. Robinson Joel,
  • G. Manikandan

摘要

Medical image enhancement plays an essential role in delivering high-quality medical images. Moreover, healthcare industries are transformed into modern healthcare, where all the medical tools are equipped with the latest computing innovations. Since, medical imaging tools are very helpful to detect diseases and asymptomatic treatment, the image enhancement approach has the ability to convert low-resolution images into high-resolution images, and it has significant importance in scientific research and clinical applications. However, some image enhancement mechanisms can oversmooth the image and lose important details about the initial image, which makes the detection and analysis process critical. Therefore, to enhance the MRI brain image an innovative model, namely Enlighten Generative MIRNet (EGMIRNet) is proposed. Initially, the MRI brain image is given as input for pre-processing, which is done based on Medav filter. Then, the pre-processed image is employed in image enhancement process, which is done by EGMIRNet, the combination of Enlighten Generative Adversarial Network (GAN) and Multiple Identities Representation Network (MIRNet). In addition, the developed EGMIRNet is evaluated based on metrics like Degree of Distortion (DD), Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), and Structural Similarity Index Measure (SSIM), and it has obtained the values of 0.056, 48.187 dB, 0.004, and 0.957.