Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May 20:27:e71560.
doi: 10.2196/71560.

Digital Biomarkers for Parkinson Disease: Bibliometric Analysis and a Scoping Review of Deep Learning for Freezing of Gait

Wenhao Qi #  1   2   3 , Shiying Shen #  1   2 , Chaoqun Dong #  1   3 , Mengjiao Zhao #  4   5 , Shuaiqi Zang  1   2 , Xiaohong Zhu  1   2   3 , Jiaqi Li  1 , Bin Wang  1 , Yankai Shi  1 , Yongze Dong  5 , Huajuan Shen  5 , Junling Kang  6 , Xiaodong Lu  2 , Guowei Jiang  7 , Jingsong Du  8 , Eryi Shu  9 , Qingbo Zhou  10   11 , Jinghua Wang  2 , Shihua Cao  1   3   12
Affiliations

Digital Biomarkers for Parkinson Disease: Bibliometric Analysis and a Scoping Review of Deep Learning for Freezing of Gait

Wenhao Qi et al. J Med Internet Res. .

Abstract

Background: With the rapid development of digital biomarkers in Parkinson disease (PD) research, it has become increasingly important to explore the current research trends and key areas of focus.

Objective: This study aimed to comprehensively evaluate the current status, hot spots, and future trends of global PD biomarker research, and provide a systematic review of deep learning models for freezing of gait (FOG) digital biomarkers.

Methods: This study used bibliometric analysis based on the Web of Science Core Collection database to conduct a comprehensive analysis of the multidimensional landscape of Parkinson digital biomarkers. After identifying research hot spots, the study also followed the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines for a scoping review of deep learning models for FOG from 5 databases: Web of Science, PubMed, IEEE Xplore, Embase, and Google Scholar.

Results: A total of 750 studies were included in the bibliometric analysis, and 40 studies were included in the scoping review. The analysis revealed a growing number of related publications, with 3700 researchers contributing. Neurology had the highest average annual participation rate (12.46/19, 66%). The United States contributed the most research (192/1171, 16.4%), with 210 participating institutions, which was the highest among all countries. In the study of deep learning models for FOG, the average accuracy of the models was 0.92, sensitivity was 0.88, specificity was 0.90, and area under the curve was 0.91. In addition, 31 (78%) studies indicated that the best models were primarily convolutional neural networks or convolutional neural networks-based architectures.

Conclusions: Research on digital biomarkers for PD is currently at a stable stage of development, with widespread global interest from countries, institutions, and researchers. However, challenges remain, including insufficient interdisciplinary and interinstitutional collaboration, as well as a lack of corporate funding for related projects. Current research trends primarily focus on motor-related studies, particularly FOG monitoring. However, deep learning models for FOG still lack external validation and standardized performance reporting. Future research will likely progress toward deeper applications of artificial intelligence, enhanced interinstitutional collaboration, comprehensive analysis of different data types, and the exploration of digital biomarkers for a broader range of Parkinson symptoms.

Trial registration: Open Science Foundation (OSF Registries) OSF.IO/RG8Y3; https://doi.org/10.17605/OSF.IO/RG8Y3.

Keywords: AI; PRISMA; Parkinson disease; artificial intelligence; bibliometric analysis; deep learning; digital biomarker; machine learning.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: None declared.

Figures

Figure 1
Figure 1
Specific framework and steps of bibliometric and scoping review research. AI: artificial intelligence; PRISMA-ScR: Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews; WoS: Web of Science.
Figure 2
Figure 2
The specific process of literature screening and research content for bibliometric and scoping review studies. REM: rapid eye movement; RQ: research question; WoS: Web of Science.
Figure 3
Figure 3
Distribution of Parkinson digital biomarker research output. (A) Annual output distribution and trend graph. (B) Change point year analysis graph.
Figure 4
Figure 4
Author analysis overview. (A) Author publication distribution. (B) Core author collaboration network map.
Figure 5
Figure 5
The participation rate of various disciplines and the collaboration network diagram. (A) Annual participation rate trends in medical-related disciplines. (B) Annual participation rate trends in engineering and other disciplines.
Figure 6
Figure 6
Participation of researchers from different disciplines. (A) Average participation rate by discipline. (B) Collaboration network among disciplines.
Figure 7
Figure 7
Overview of institution analysis. (A) Institutional collaboration network map. (B) Heatmap of collaboration intensity matrix between different types of institutions.
Figure 8
Figure 8
Country analysis. (A) Global distribution map of Parkinson digital biomarker research output by country. (B) Annual output distribution trends of high-producing countries. (C) International collaboration Sankey diagram of Parkinson digital biomarker research by country.
Figure 9
Figure 9
Number of institutions and distribution by type in high-producing countries.
Figure 10
Figure 10
Funding project analysis overview. (A) Distribution of funding numbers by type. (B) Distribution of research output by funding type. (C) Distribution of top 10 funding projects by number of instances. (D) Distribution of disciplines funded by top 10 funding projects. JSPS: Japan Society for the Promotion of Science; UK: United Kingdom.
Figure 11
Figure 11
Disciplinary analysis clusters and matrix. (A) Clustering results based on Web of Science major disciplines. (B) Clustering results based on Web of Science subdisciplines.
Figure 12
Figure 12
Journal and highly cited literature analysis. (A) Co-cited journal clustering map. (B) High-output journal distribution map. (C) Highly cited literature distribution map.
Figure 13
Figure 13
Keyword analysis. (A) Keyword theme trend analysis. (B) Keyword burst distribution.
Figure 14
Figure 14
High-frequency keyword clustering. (A) High-frequency keyword cluster hill diagram. (B) High-frequency keyword hierarchical clustering diagram.
Figure 15
Figure 15
Theme trend changes. (A) Annual theme trend changes. (B) Development of Parkinson disease digital biomarkers in technology, symptoms, and devices. REM: rapid eye movement.
Figure 16
Figure 16
The entire process of constructing a deep learning model for Parkinson freezing of gait (FOG): (A) the occurrence of FOG; (B) selection of digital devices; (C) data collection and analysis from digital devices; (D) application of deep learning and neural networks; (E) training of the model and output of results.
Figure 17
Figure 17
Geographic and time-zone distribution of the output of deep learning models for Parkinson freezing of gait.
Figure 18
Figure 18
Performance distribution of deep learning models for freezing of gait in Parkinson disease. ACC: accuracy; AUC: area under the curve; SEN: sensitivity; SPE: specificity.
Figure 19
Figure 19
Future development directions in Parkinson digital biomarker research. REM: rapid eye movement.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Stoessl AJ, Martin WW, McKeown MJ, Sossi V. Advances in imaging in Parkinson's disease. Lancet Neurol. 2011 Nov;10(11):987–1001. doi: 10.1016/S1474-4422(11)70214-9.S1474-4422(11)70214-9 - DOI - PubMed
    1. Selby G. Treatment of parkinsonism. Drugs. 1976;11(1):61–70. doi: 10.2165/00003495-197611010-00006. - DOI - PubMed
    1. Tolosa E, Garrido A, Scholz SW, Poewe W. Challenges in the diagnosis of Parkinson's disease. Lancet Neurol. 2021 May;20(5):385–97. doi: 10.1016/S1474-4422(21)00030-2. https://europepmc.org/abstract/MED/33894193 S1474-4422(21)00030-2 - DOI - PMC - PubMed
    1. Sieberts SK, Schaff J, Duda M, Pataki B, Sun M, Snyder P, Daneault J, Parisi F, Costante G, Rubin U, Banda P, Chae Y, Chaibub Neto E, Dorsey ER, Aydın Z, Chen A, Elo LL, Espino C, Glaab E, Goan E, Golabchi FN, Görmez Y, Jaakkola MK, Jonnagaddala J, Klén R, Li D, McDaniel C, Perrin D, Perumal TM, Rad NM, Rainaldi E, Sapienza S, Schwab P, Shokhirev N, Venäläinen MS, Vergara-Diaz G, Zhang Y, Parkinson’s Disease Digital Biomarker Challenge Consortium. Wang Y, Guan Y, Brunner D, Bonato P, Mangravite LM, Omberg L. Crowdsourcing digital health measures to predict Parkinson's disease severity: the Parkinson's Disease Digital Biomarker DREAM Challenge. NPJ Digit Med. 2021 Mar 19;4(1):53. doi: 10.1038/s41746-021-00414-7. https://doi.org/10.1038/s41746-021-00414-7 10.1038/s41746-021-00414-7 - DOI - DOI - PMC - PubMed
    1. Kalia LV, Lang AE. Parkinson's disease. Lancet. 2015 Aug 29;386(9996):896–912. doi: 10.1016/S0140-6736(14)61393-3.S0140-6736(14)61393-3 - DOI - PubMed

Publication types

LinkOut - more resources