Scholarlink

Globally kidney cancer, renal cell carcinoma (RCC), and other malignancies of the system in renal pose a significant health challenge. For effective treatment, accurate detection of kidney tumors is crucial to improve patient outcomes. Recent advancement in medical imaging techniques and deep learning algorithms has led to the development of computer-aided testing systems, which are effective tools for automating the segmentation and categorization of kidney tumors. Initially, essential images for the evaluation are garnered from the benchmark sources. Then, image pre-processing is conducted in the input image using the weighted mean histogram equalization technique and the contrast enhancement method. Next, the pre-processed images are forwarded to the segmentation region. Here, the InceptionResNetV2 technique is utilized and their parameters are tuned by the Improved Sea Lion Optimization (ISLO) algorithm. Further, the segmented images are provided as the input to the kidney tumor classification stage. Here, the implemented model named 3D U-Net is utilized to classify the kidney tumor effectively and accurately.

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

ISLO-Tuned InceptionResNetV2 and 3D U-Net: A Powerful Duo for Automated Renal Cancer Diagnosis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Vinitkumar Vasantbhai Patel,
  • Arvind R. Yadav

摘要

Globally kidney cancer, renal cell carcinoma (RCC), and other malignancies of the system in renal pose a significant health challenge. For effective treatment, accurate detection of kidney tumors is crucial to improve patient outcomes. Recent advancement in medical imaging techniques and deep learning algorithms has led to the development of computer-aided testing systems, which are effective tools for automating the segmentation and categorization of kidney tumors. Initially, essential images for the evaluation are garnered from the benchmark sources. Then, image pre-processing is conducted in the input image using the weighted mean histogram equalization technique and the contrast enhancement method. Next, the pre-processed images are forwarded to the segmentation region. Here, the InceptionResNetV2 technique is utilized and their parameters are tuned by the Improved Sea Lion Optimization (ISLO) algorithm. Further, the segmented images are provided as the input to the kidney tumor classification stage. Here, the implemented model named 3D U-Net is utilized to classify the kidney tumor effectively and accurately.