Georgios Bouloukakis, Georgios Karamitros, Gregory A. Lamaris, Wesley P. Thayer, Galen Perdikis, Feng Zhang, William C. Lineaweaver
Abstract
Artificial intelligence (AI)-assisted approaches may allow surgical research trends to be analyzed at scale and projected over time. However, their use in forecasting the evolution of microsurgical scholarship remains limited. This study developed an AI-assisted bibliometric framework to characterize and project global clinical and experimental microsurgery publication trends.
Introduction
The next frontier in surgical science lies not only in measuring what has already been achieved, but also in understanding where scientific activity is likely to move next [1]. In parallel with its expanding role in diagnostics, operative planning, imaging, and clinical decision support, ar-tificial intelligence (AI) is increasingly being applied to the study of biomedical knowledge itself [1–3]. Within academic surgery, this creates an opportunity for bibliometric analysis to evolve from retrospective mapping toward more dynamic, forward-looking models of scholarly activity [4,5]
Methods
Study design and conceptual framework
This study developed an AI-assisted forecasting model to analyze and predict the growth of clinical and experimental microsurgical publications from 2010 to 2024, with projections through 2030. The analytic framework integrates automated data extraction, contextual classification, and pre-dictive modeling to generate a temporally dynamic and geographically resolved map of global publication activity [4,17]. T
Results
Dataset overview and AI system performance
The AI-assisted forecasting framework processed 90,902 records published between 2010 and 2024 across twenty microsurgery-related journals. After removal of 7,769 incomplete entries (missing metadata or affiliation), 83,133 articles underwent automated text-mining and contextual classification. The automated workflow excluded 71,355 non-microsurgical records, producing a validated dataset of 11,561 publications with verifiable first-author country attribution.
Discussion
This study demonstrates how an AI-assisted bibliometric workflow can support large-scale assessment of microsurgical publication activity. By integrating automated metadata extraction, contextual keyword-based classification, human-reviewed validation, and conventional statistical forecasting, the framework provides a reproducible approach for examining historical and projected publication trends across time, geography, authorship, and research domains.
Conclusion
This study presents an AI-assisted bibliometric workflow for characterizing and projecting micro-surgical publication trends. By integrating automated PubMed metadata extraction, rule-based contextual keyword classification, human-reviewed validation, and conventional statistical forecasting, the workflow provides a reproducible approach for assessing publication activity across time, geography, authorship, and thematic domains.
Acknowledgments
Disclosure: The authors have no financial interest to declare in relation to the content of this study.
Citation: Bouloukakis G, Karamitros G, Lamaris GA, Thayer WP, Perdikis G, Zhang F, et al. (2026) AI-assisted forecasting in microsurgery: A dual-component framework for global publication trends. PLoS One 21(8): e0357186. https://doi.org/10.1371/journal.pone.0357186
Editor: Xiaoen Wei, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, CHINA
Received: May 26, 2026; Accepted: August 13, 2026; Published: August 28, 2026
Copyright: © 2026 Bouloukakis et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The forecasting algorithm used in this study has been deposited in a public GitHub repository and archived on Zenodo to facilitate transparency and reproducibility. The repository contains a standalone Python implementation of the model comparison and forecasting workflow, including linear regression, quadratic polynomial regression, ARIMA, and Holt’s exponential smoothing, together with synthetic example data for reproducibility testing. To protect manuscript-sensitive study results and article-level metadata, the public repository does not include raw datasets, author information, affiliations, abstracts, or manuscript-derived aggregate results. Repository: https://github.com/byorgos/microsurgery-forecast Archived release: https://doi.org/10.5281/zenodo.20396105 Version: v0.1.0 License: MIT.
Funding: The author(s) received no specific funding for this work.
Competing interests: he authors have declared that no competing interests exist.
