A bibliometric analysis of worldwide cancer research using machine learning methods

Area of Science:

• Computational biology
• Bioinformatics
• Medical informatics

Background:

• Machine learning (ML) applications in cancer research are expanding due to advancements in computer technology.
• Understanding the current landscape, trends, and collaborations is crucial for future progress.

Purpose of the Study:

• To conduct a bibliometric analysis of ML in cancer research from 2011-2021.
• To identify research status, trends, topics, collaborations, and future directions.

Main Methods:

• Bibliometric analysis of 6206 PubMed articles (2011-2021).
• Utilized Python for bibliometrics, Gephi for social network analysis, and Latent Dirichlet Allocation for topic modeling.

Main Results:

• The field shows vigorous development and increasing impact, with significant innovation at the ML-cancer intersection.
• Nature Communications (influential) and Scientific Reports (prolific) are key journals.
• The United States and Harvard University lead contributions.
• Core research topics include Support Vector Machine, classification, and deep learning.

Conclusions:

• The study provides insights into the dynamic development and trends of ML in cancer research.
• Findings aid researchers and practitioners in understanding research hotspots and potential future directions.