Scholarlink

Dermatologists can discover skin malignancies in their early stages and save patients’ lives by evaluating skin lesions. Skin lesions can occur for a variety of reasons, including allergies, infections, sun exposure, and other diseases. Because of the high degree of similarity among the many types of skin lesions, an incorrect diagnosis is made, and clinical evaluation of skin lesions is hampered by lengthy turnaround times. Dermatologists may now classify skin lesions with the help of deep learning techniques. The skin lesions in this study were classified into seven classes using convolutional neural network (CNN): melanocytic nevi, melanoma, melanoma-like lesions, melanoma-like lesions, Benign keratosis-like lesions, Basal cell carcinoma, Actinic keratoses, Vascular lesions, and Dermatofibroma. This endeavor in dermatology demonstrates how ML and DL approaches can aid clinical decision-making and improve patient outcomes. Utilising 10015 images of skin lesions from the publically available HAM10000 dataset, we test the effectiveness of the suggested model using a variety of criteria, such as accuracy and precision.

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

CNN Approach for Skin Lesion Classification Using Dermoscopic Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • K. V. Raghavender,
  • Gummadivelli Maneesha,
  • Indoori Vaishnavi,
  • Nalagati Tejaswini,
  • Jupaka Sharanya

摘要

Dermatologists can discover skin malignancies in their early stages and save patients’ lives by evaluating skin lesions. Skin lesions can occur for a variety of reasons, including allergies, infections, sun exposure, and other diseases. Because of the high degree of similarity among the many types of skin lesions, an incorrect diagnosis is made, and clinical evaluation of skin lesions is hampered by lengthy turnaround times. Dermatologists may now classify skin lesions with the help of deep learning techniques. The skin lesions in this study were classified into seven classes using convolutional neural network (CNN): melanocytic nevi, melanoma, melanoma-like lesions, melanoma-like lesions, Benign keratosis-like lesions, Basal cell carcinoma, Actinic keratoses, Vascular lesions, and Dermatofibroma. This endeavor in dermatology demonstrates how ML and DL approaches can aid clinical decision-making and improve patient outcomes. Utilising 10015 images of skin lesions from the publically available HAM10000 dataset, we test the effectiveness of the suggested model using a variety of criteria, such as accuracy and precision.