Scholarlink

<p>Skin screening apps based on artificial intelligence are becoming increasingly relevant for the early detection of skin cancer. Convolutional neural networks achieve accuracy comparable to dermatologists. The use of explainable artificial intelligence aims to enhance clinicians’ trust in AI-assisted diagnoses. Studies reveal methodological limitations: training datasets are often of low quality, algorithms lack transparency, and challenges exist regarding bias and generalizability. While skin screening apps can support diagnosis, particularly in primary care, a&#xa0;full replacement of dermatological expertise remains unlikely. Clinical implementation requires clear regulatory frameworks, interoperability within healthcare systems, and standardized quality assurance. Future developments may integrate AI with teledermatology and wearable technologies to further improve diagnostic accuracy and accessibility.</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

Haut-Screening-Apps in der klinischen Praxis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月27日

查看全文
  • Felix Erhard Gerschewski

摘要

Skin screening apps based on artificial intelligence are becoming increasingly relevant for the early detection of skin cancer. Convolutional neural networks achieve accuracy comparable to dermatologists. The use of explainable artificial intelligence aims to enhance clinicians’ trust in AI-assisted diagnoses. Studies reveal methodological limitations: training datasets are often of low quality, algorithms lack transparency, and challenges exist regarding bias and generalizability. While skin screening apps can support diagnosis, particularly in primary care, a full replacement of dermatological expertise remains unlikely. Clinical implementation requires clear regulatory frameworks, interoperability within healthcare systems, and standardized quality assurance. Future developments may integrate AI with teledermatology and wearable technologies to further improve diagnostic accuracy and accessibility.