Evolution of knee osteoarthritis biomechanics research, 2005–2025: bibliometric mapping and machine-learning-assisted topic modeling
4 affiliations
- 1School of Acupuncture and Massage, Anhui University of Chinese Medicine, Hefei, Anhui, China
- 2College of Integrated Chinese and Western Medicine, Anhui University of Chinese Medicine, Hefei, Anhui, China
- 3Functional Activity and Resource Utilization of Edible and Medicinal Fungi Anhui Provincial Joint Key Laboratory, Anhui University of Chinese Medicine, Hefei, Anhui, China
- 4Anhui Provincial Key Laboratory of Tumor Evolution and Intelligent Diagnosis and Treatment, Bengbu, Anhui, China
Abstract
BackgroundKnee osteoarthritis (KOA) involves interactions among whole-joint loading, tissue mechanics, and cellular responses. This study mapped the structure and temporal evolution of KOA biomechanics research using bibliometric analysis and machine-learning-assisted topic modeling.MethodsEnglish-language articles and reviews published from 2005 to 2025 were retrieved from the Web of Science Core Collection and Scopus, merged, deduplicated, and screened. The final corpus comprised 3110 publications. Publication patterns, collaboration networks, citation structures, and keyword themes were evaluated, and non-negative matrix factorization was applied to bibliographic text to identify latent topics and their temporal distribution.ResultsAnnual output increased from 37 publications in 2005 to 321 in 2025. The United States led in publication output and total citations, while the University of Melbourne was the most productive institution. Keyword analysis identified five major domains spanning mechanobiology, gait and neuromuscular control, conservative load-modifying interventions, alignment correction and reconstruction, and meniscal and ligament mechanics. The selected model comprised 19 topics. Gait and lower-limb kinematics had the largest overall share. The greatest increases in mean annual topic share from 2016-2020 to 2021-2025 involved stem cells and regenerative cartilage repair, chondrocyte mechanotransduction, and meniscal tears, extrusion, and root repair. Recent changes in topic share were strongly correlated between hard and normalized soft assignments (Spearman's ρ = 0.926).ConclusionOverall, publication output increased, and topic-share patterns indicated greater relative representation of mechanobiology, regenerative repair, meniscal preservation, and alignment correction, while gait and loading research remained prominent. These findings provide a descriptive map of the field's development and changing research priorities.
Related
- Functional Brain Imaging in Knee Osteoarthritis: A Bibliometric Analysis of Research Trends and Emerging Clinical Perspectives
- Global research hotspots and trends of osteoporosis and osteoarthritis comorbidity from 2005 to 2025: a bibliometric and visualized analysis
- Frontiers and hotspots evolution of machine learning in gout from 2004–2023 : A bibliometric analysis
- A bibliometric analysis of trace elements and diabetic kidney disease: scientific hotspots and research evolution.
- Research Trends, Knowledge Evolution and Pharmacological Advances of Cordyceps cicadae: A Bibliometric Analysis
BibTeX
@article{lu2026evolution,
title = {{Evolution of knee osteoarthritis biomechanics research, 2005–2025: bibliometric mapping and machine-learning-assisted topic modeling}},
author = {Fei Lu and Lu Liu and Peng Yang and Hang Gao and Hang Song and Hang Song and Hang Song and Guoquan Wang and Zongbao Wang},
journal = {Frontiers in Medicine},
year = {2026},
month = oct,
doi = {10.3389/fmed.2026.1968099},
url = {https://doi.org/10.3389/fmed.2026.1968099},
note = {CC BY},
}
RIS
TY - JOUR
TI - Evolution of knee osteoarthritis biomechanics research, 2005–2025: bibliometric mapping and machine-learning-assisted topic modeling
AU - Fei Lu
AU - Lu Liu
AU - Peng Yang
AU - Hang Gao
AU - Hang Song
AU - Hang Song
AU - Hang Song
AU - Guoquan Wang
AU - Zongbao Wang
PY - 2026
DA - 2026/10/26
JO - Frontiers in Medicine
DO - 10.3389/fmed.2026.1968099
UR - https://doi.org/10.3389/fmed.2026.1968099
AB - BackgroundKnee osteoarthritis (KOA) involves interactions among whole-joint loading, tissue mechanics, and cellular responses. This study mapped the structure and temporal evolution of KOA biomechanics research using bibliometric analysis and machine-learning-assisted topic modeling.MethodsEnglish-language articles and reviews published from 2005 to 2025 were retrieved from the Web of Science Core Collection and Scopus, merged, deduplicated, and screened. The final corpus comprised 3110 publications. Publication patterns, collaboration networks, citation structures, and keyword themes were evaluated, and non-negative matrix factorization was applied to bibliographic text to identify latent topics and their temporal distribution.ResultsAnnual output increased from 37 publications in 2005 to 321 in 2025. The United States led in publication output and total citations, while the University of Melbourne was the most productive institution. Keyword analysis identified five major domains spanning mechanobiology, gait and neuromuscular control, conservative load-modifying interventions, alignment correction and reconstruction, and meniscal and ligament mechanics. The selected model comprised 19 topics. Gait and lower-limb kinematics had the largest overall share. The greatest increases in mean annual topic share from 2016-2020 to 2021-2025 involved stem cells and regenerative cartilage repair, chondrocyte mechanotransduction, and meniscal tears, extrusion, and root repair. Recent changes in topic share were strongly correlated between hard and normalized soft assignments (Spearman's ρ = 0.926).ConclusionOverall, publication output increased, and topic-share patterns indicated greater relative representation of mechanobiology, regenerative repair, meniscal preservation, and alignment correction, while gait and loading research remained prominent. These findings provide a descriptive map of the field's development and changing research priorities.
SN - 2296-858X
ER -