Scholarlink

Background <p>AI is changing how we do things in Dermatology Oncology, especially with regards to finding and classifying skin cancers by looking at images of skin. The number of articles published about AI and skin cancer have grown rapidly over the last ten years, creating a very large and very interdisciplinary body of work, which requires a systematic bibliometric analysis in order to understand the trend of the work, how the themes evolve, the collaborations that exist among researchers, and what are the most important sources of information.</p> Methods <p>A bibliometric study of research in this area was undertaken based on a search of the Web of Science Core Collection for articles and reviews published between 2010 and 2024. The results from the bibliographic data set were then visualized by employing VOSviewer to identify clusters of words (keyword co-occurrence) that appear together in abstracts and titles, visualize how these clusters have developed temporally (temporal overlay), map collaboration among countries (country-level collaboration), and identify clusters of co-cited sources (source co-citation).</p> Results <p>The study found that there has been an important increase of the number of publications per year, especially from around 2015 onwards. Dermoscopy and melanoma have become the key clinical themes; whereas, deep learning and convolutional neural networks have been the mainstay of the methodological developments in the last few years. International collaborative networks were very well connected and the United States, China and India were the largest hubs. A source co-citation analysis found a strong inter-disciplinary knowledge base across dermatology, medical imaging and artificial intelligence, and oncology.</p> Conclusions <p>AI research in skin cancer has evolved rapidly toward data-driven, deep learning–based approaches with increasing clinical relevance. Bibliometric findings highlight the maturation of the field, its global scope, and the growing integration of methodological innovation with dermatologic oncology practice. Level of Evidence: not gradable.</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

Artificial intelligence in skin cancer research (2010–2024): a bibliometric analysis using the web of science database

  • 期刊论文

  • 发表日期：

    2026年5月29日

查看全文
  • Mohammed Naif Alsubhi,
  • Faris Alaa Sultan,
  • Abdullah Ahmed Alabbasi,
  • Muhanad Saleh Alzahrani,
  • Maha S Alqahtani,
  • Muhannad Q. Alqirnas

摘要

Background

AI is changing how we do things in Dermatology Oncology, especially with regards to finding and classifying skin cancers by looking at images of skin. The number of articles published about AI and skin cancer have grown rapidly over the last ten years, creating a very large and very interdisciplinary body of work, which requires a systematic bibliometric analysis in order to understand the trend of the work, how the themes evolve, the collaborations that exist among researchers, and what are the most important sources of information.

Methods

A bibliometric study of research in this area was undertaken based on a search of the Web of Science Core Collection for articles and reviews published between 2010 and 2024. The results from the bibliographic data set were then visualized by employing VOSviewer to identify clusters of words (keyword co-occurrence) that appear together in abstracts and titles, visualize how these clusters have developed temporally (temporal overlay), map collaboration among countries (country-level collaboration), and identify clusters of co-cited sources (source co-citation).

Results

The study found that there has been an important increase of the number of publications per year, especially from around 2015 onwards. Dermoscopy and melanoma have become the key clinical themes; whereas, deep learning and convolutional neural networks have been the mainstay of the methodological developments in the last few years. International collaborative networks were very well connected and the United States, China and India were the largest hubs. A source co-citation analysis found a strong inter-disciplinary knowledge base across dermatology, medical imaging and artificial intelligence, and oncology.

Conclusions

AI research in skin cancer has evolved rapidly toward data-driven, deep learning–based approaches with increasing clinical relevance. Bibliometric findings highlight the maturation of the field, its global scope, and the growing integration of methodological innovation with dermatologic oncology practice. Level of Evidence: not gradable.