Scholarlink

Identification of skin cancer in its early stages can increase the chances of successful treatment and this is so because skin cancer is one of the most prevalent cancers worldwide. Recent advancements in deep learning have resulted in promising methods for automated skin cancer identification using dermoscopic images, with the potential for boosting diagnostic accuracy and efficiency. This article introduces a deep convolutional neural network model that uses Keras’ Sequential API to categorize skin lesions from the HAM10000 dataset. The model is built as follows, comprising sequential Conv2D and MaxPooling2D layers for features extraction followed by densely connected ReLU and Softmax neurons for classification purpose and efficacy. The model achieved a high test accuracy of 97%, demonstrating that it can differentiate between different types of skin lesions. This work offers an efficient way for the automatic detection of skin cancer, one that can be integrated into clinical diagnosis support systems.

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

Deep Learning for Dermoscopic Diagnosis: High-Accuracy CNN in Skin Cancer Classification

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Greeshma Arya,
  • Saloni Choudhary,
  • Tripti Verma,
  • Janvi

摘要

Identification of skin cancer in its early stages can increase the chances of successful treatment and this is so because skin cancer is one of the most prevalent cancers worldwide. Recent advancements in deep learning have resulted in promising methods for automated skin cancer identification using dermoscopic images, with the potential for boosting diagnostic accuracy and efficiency. This article introduces a deep convolutional neural network model that uses Keras’ Sequential API to categorize skin lesions from the HAM10000 dataset. The model is built as follows, comprising sequential Conv2D and MaxPooling2D layers for features extraction followed by densely connected ReLU and Softmax neurons for classification purpose and efficacy. The model achieved a high test accuracy of 97%, demonstrating that it can differentiate between different types of skin lesions. This work offers an efficient way for the automatic detection of skin cancer, one that can be integrated into clinical diagnosis support systems.