Tumor Image Segmentation: A Bibliometric Analysis from 2003 to 2024.

Area of Science:

• Medical Imaging Analysis
• Bibliometrics
• Artificial Intelligence in Oncology

Background:

• Bibliometric analysis is a valuable tool for understanding research trends and hotspots in scientific fields.
• Tumor image segmentation is a critical area of study with significant implications for disease diagnosis and treatment.

Purpose of the Study:

• To conduct a bibliometric analysis of tumor image segmentation research.
• To identify current research hotspots, trends, and emerging topics in tumor image segmentation.
• To provide research guidelines for scholars in the field.

Main Methods:

• Bibliometric analysis of 3377 articles published between 2003 and 2024, sourced from the Web of Science database.
• Analysis of publication volume, country/region, institution, journal, author, keywords, and references.
• Visualization of co-authorship, co-citation, and co-occurrence using VOSviewer.

Main Results:

• A significant increase in publications since 2016, with 576 articles in 2023.
• Mainland China leads in publication volume, while IEEE Transactions on Medical Imaging is the most cited journal.
• Key research clusters identified: segmentation methods, applications, CT-based segmentation, and MRI-based segmentation, with Transformer, Attention Mechanism, and U-Net as emerging keywords.

Conclusions:

• Tumor image segmentation research has grown steadily, with rapid advancements in methods like U-Net and MAMBA.
• These advancements are crucial for improving disease prevention and monitoring.
• The study highlights global collaboration and provides insights into research trends and future directions.