CoChat

Alzheimer's disease digital biomarkers multidimensional landscape and AI model scoping review

Open access

The DOI matches its Crossref record; 207 of the 207 references it cites were found in a registry or an index.

Abstract

As digital biomarkers gain traction in Alzheimer's disease (AD) diagnosis, understanding recent advancements is crucial. This review conducts a bibliometric analysis of 431 studies from five online databases: Web of Science, PubMed, Embase, IEEE Xplore, and CINAHL, and provides a scoping review of 86 artificial intelligence (AI) models. Research in this field is supported by 224 grants across 54 disciplines and 1403 institutions in 44 countries, with 2571 contributing researchers. Key focuses include motor activity, neurocognitive tests, eye tracking, and speech analysis. Classical machine learning models dominate AI research, though many lack performance reporting. Of 21 AD-focused models, the average AUC is 0.887, while 45 models for mild cognitive impairment show an average AUC of 0.821. Notably, only 2 studies incorporated external validation, and 3 studies performed model calibration. This review highlights the progress and challenges of integrating digital biomarkers into clinical practice.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41746-025-01640-z

Cites

207 of 207 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Literature retrieval and inclusion process flowchart. a Flow diagram illustrating the PRISMA approach for the identification, screening, and selection of studies. b Specific research content.
Fig. 1 Literature retrieval and inclusion process flowchart. a Flow diagram illustrating the PRISMA approach for the identification, screening, and selection of studies. b Specific research content.CC BY-NC-ND · PubMed Central
Fig. 2 Distribution and Trends in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Publication output distribution and trends over time. The red dashed line represents the trend line (aTrend line:y = 0.2297x2–1.775x + 3.8271). b Phases of publication output in Alzheimer’s disease digital biomarkers diagnostic research (aTrend line:y = 0.2297x2–1.775x + 3.8271). Asterisk indicates that the slope is significantly different from zero at the α = 0.05 level. Final selected model: 3 joinpoints.
Fig. 2 Distribution and Trends in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Publication output distribution and trends over time. The red dashed line represents the trend line (aTrend line:y = 0.2297x2–1.775x + 3.8271). b Phases of publication output in Alzheimer’s disease digital biomarkers diagnostic research (aTrend line:y = 0.2297x2–1.775x + 3.8271). Asterisk indicates that the slope is significantly different from zero at the α = 0.05 level. Final selected model: 3 joinpoints.CC BY-NC-ND · PubMed Central
Fig. 3 Institutional Collaboration in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Distribution of institution types involved in Alzheimer’s disease digital biomarkers diagnostic research. b Output by institution type in Alzheimer’s disease digital biomarkers diagnostic research. c Network diagram of institutional collaboration. Color coding is used to display clusters, with institutions within the same cluster sharing the same color. The size of the circles increases with the number of publications. d Evolution of institutional collaborations over time. Color coding is used to represent the average time for constructing institutional collaboration networks. The size of the circles increases with the number of publications.
Fig. 3 Institutional Collaboration in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Distribution of institution types involved in Alzheimer’s disease digital biomarkers diagnostic research. b Output by institution type in Alzheimer’s disease digital biomarkers diagnostic research. c Network diagram of institutional collaboration. Color coding is used to display clusters, with institutions within the same cluster sharing the same color. The size of the circles increases with the number of publications. d Evolution of institutional collaborations over time. Color coding is used to represent the average time for constructing institutional collaboration networks. The size of the circles increases with the number of publications.CC BY-NC-ND · PubMed Central
Fig. 4 Productivity and Collaboration Among Nations in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Temporal distribution of output by high-producing countries. b Chord diagram of international collaborations among countries. c Geographic distribution map of publication by country in Alzheimer’s disease digital biomarkers diagnostic research. The colors representing countries/regions have no specific meaning; only the thickness of the lines between them is significant, indicating the frequency of collaboration between different countries. The thickness of the lines corresponds to the values on their respective axes.
Fig. 4 Productivity and Collaboration Among Nations in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Temporal distribution of output by high-producing countries. b Chord diagram of international collaborations among countries. c Geographic distribution map of publication by country in Alzheimer’s disease digital biomarkers diagnostic research. The colors representing countries/regions have no specific meaning; only the thickness of the lines between them is significant, indicating the frequency of collaboration between different countries. The thickness of the lines corresponds to the values on their respective axes.CC BY-NC-ND · PubMed Central
Fig. 5 Heatmaps of Disciplinary Publications in Alzheimer’s Disease Digital Biomarkers Diagnostic Research Over Three Time Periods. a 2004–2012. b 2013–2018. c 2019–2024. Color indicates the popularity of a discipline during the given time period, with redder colors representing higher popularity and dominance.
Fig. 5 Heatmaps of Disciplinary Publications in Alzheimer’s Disease Digital Biomarkers Diagnostic Research Over Three Time Periods. a 2004–2012. b 2013–2018. c 2019–2024. Color indicates the popularity of a discipline during the given time period, with redder colors representing higher popularity and dominance.CC BY-NC-ND · PubMed Central
Fig. 6 Temporal Patterns of Disciplinary Publications in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. The different colors of the bars represent different disciplines, and the length of each bar indicates the proportion of output from a particular discipline during the specified period.
Fig. 6 Temporal Patterns of Disciplinary Publications in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. The different colors of the bars represent different disciplines, and the length of each bar indicates the proportion of output from a particular discipline during the specified period.CC BY-NC-ND · PubMed Central
Fig. 7 Types and Quantities of Funding in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Funding trends in Alzheimer’s disease digital biomarkers diagnostic research. The red dashed line represents the trend line (aTrend line:y = 0.7712x2–6.0744x + 9.2722), (b) Distribution of funding types. c Quantities of each type of funding.
Fig. 7 Types and Quantities of Funding in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Funding trends in Alzheimer’s disease digital biomarkers diagnostic research. The red dashed line represents the trend line (aTrend line:y = 0.7712x2–6.0744x + 9.2722), (b) Distribution of funding types. c Quantities of each type of funding.CC BY-NC-ND · PubMed Central
Fig. 8 Keywords Analysis in Alzheimer’s disease Digital Biomarkers Diagnostic Research. a Temporal heatmap of high-frequency keywords. Color represents the proportion of keyword frequency for that year relative to the total frequency of the keyword. The more frequent the keyword, the redder the color, indicating a more mature topic. b Clustering diagram of keywords co-occurrence. Color coding is used to display clusters, with keywords within the same cluster sharing the same color.
Fig. 8 Keywords Analysis in Alzheimer’s disease Digital Biomarkers Diagnostic Research. a Temporal heatmap of high-frequency keywords. Color represents the proportion of keyword frequency for that year relative to the total frequency of the keyword. The more frequent the keyword, the redder the color, indicating a more mature topic. b Clustering diagram of keywords co-occurrence. Color coding is used to display clusters, with keywords within the same cluster sharing the same color.CC BY-NC-ND · PubMed Central
Fig. 9 Production, Distribution and Trends of Various Types of Alzheimer’s Disease Digital Biomarkers. The pie chart is used to represent the proportion of research on different types of digital biomarkers, while the line chart illustrates the trend of changes in various digital biomarkers over time.
Fig. 9 Production, Distribution and Trends of Various Types of Alzheimer’s Disease Digital Biomarkers. The pie chart is used to represent the proportion of research on different types of digital biomarkers, while the line chart illustrates the trend of changes in various digital biomarkers over time.CC BY-NC-ND · PubMed Central
Fig. 10 Sample Size Distribution in Alzheimer’s Disease Digital Biomarkers Research. The violin plot displays the distribution of sample sizes for research on each type of digital biomarker. The different colors of the violins represent distinct categories of digital biomarker research. The body of the violin represents the primary distribution range of the research, with wider sections indicating a higher number of studies and narrower sections representing fewer studies.
Fig. 10 Sample Size Distribution in Alzheimer’s Disease Digital Biomarkers Research. The violin plot displays the distribution of sample sizes for research on each type of digital biomarker. The different colors of the violins represent distinct categories of digital biomarker research. The body of the violin represents the primary distribution range of the research, with wider sections indicating a higher number of studies and narrower sections representing fewer studies.CC BY-NC-ND · PubMed Central
Fig. 11 A sankey diagram of the devices and paradigms commonly used in the collection of different digital biomarkers. The left column shows the data collection paradigms, the middle column represents different types of digital biomarkers, and the right column lists the devices used for data collection.
Fig. 11 A sankey diagram of the devices and paradigms commonly used in the collection of different digital biomarkers. The left column shows the data collection paradigms, the middle column represents different types of digital biomarkers, and the right column lists the devices used for data collection.CC BY-NC-ND · PubMed Central
Fig. 12 Core Author Collaboration in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Network diagram of core author collaboration. Color coding is used to display clusters, with researchers within the same cluster sharing the same color. The size of the circles increases with the number of publications. b Average timeline of core author collaboration initiatives. Color coding is used to represent the average time for constructing researcher collaboration networks. The size of the circles increases with the number of publications.
Fig. 12 Core Author Collaboration in Alzheimer’s Disease Digital Biomarkers Diagnostic Research. a Network diagram of core author collaboration. Color coding is used to display clusters, with researchers within the same cluster sharing the same color. The size of the circles increases with the number of publications. b Average timeline of core author collaboration initiatives. Color coding is used to represent the average time for constructing researcher collaboration networks. The size of the circles increases with the number of publications.CC BY-NC-ND · PubMed Central
Fig. 13 Participation Rate of Disciplinary Backgrounds in Alzheimer’s Disease Digital Biomarkers Research. a Annual participation rate changes in medical-related disciplines. b Annual participation rate changes in engineering-related disciplines.
Fig. 13 Participation Rate of Disciplinary Backgrounds in Alzheimer’s Disease Digital Biomarkers Research. a Annual participation rate changes in medical-related disciplines. b Annual participation rate changes in engineering-related disciplines.CC BY-NC-ND · PubMed Central
Fig. 14 Distribution of participants by disciplinary backgrounds in various types of Alzheimer’s disease digital biomarkers research. LM limb movement, EM eye movement, TM Test on mobile or ICT devices, SM Speech markers, ND Natural driving, HA Home activity, UL non-dedicated ICT biomarkers, SP Sleep pattern, BP Biofeedback or physiological signal, Other Other biomarkers, Multiple Mutiple biomarkers.) The size of the circles represents the frequency, with larger circles indicating higher frequencies. Different colors represent different disciplines.
Fig. 14 Distribution of participants by disciplinary backgrounds in various types of Alzheimer’s disease digital biomarkers research. LM limb movement, EM eye movement, TM Test on mobile or ICT devices, SM Speech markers, ND Natural driving, HA Home activity, UL non-dedicated ICT biomarkers, SP Sleep pattern, BP Biofeedback or physiological signal, Other Other biomarkers, Multiple Mutiple biomarkers.) The size of the circles represents the frequency, with larger circles indicating higher frequencies. Different colors represent different disciplines.CC BY-NC-ND · PubMed Central
Fig. 15 Heatmap of Interdisciplinary Collaboration Among Disciplines. The size and color of the sector areas represent the strength of collaboration between different disciplines. The redder the color and the larger the sector area, the stronger the collaboration between disciplines.
Fig. 15 Heatmap of Interdisciplinary Collaboration Among Disciplines. The size and color of the sector areas represent the strength of collaboration between different disciplines. The redder the color and the larger the sector area, the stronger the collaboration between disciplines.CC BY-NC-ND · PubMed Central
Fig. 16 The Process of Developing AI Models for Alzheimer’s Disease Digital Biomarkers. a Recruiting Alzheimer’s disease patient samples to form the study cohort. b Collecting data from multiple devices, including wearables, smartphones, and others. c Using sensors, mobile applications, and other tools to acquire various digital biomarker data. d Extracting and analyzing feature data such as movement, speech, eye movement, and physiological signals. e Selecting appropriate machine learning and AI algorithms to train predictive models. f Applying the models for disease classification and prediction, aiding in early detection and management.
Fig. 16 The Process of Developing AI Models for Alzheimer’s Disease Digital Biomarkers. a Recruiting Alzheimer’s disease patient samples to form the study cohort. b Collecting data from multiple devices, including wearables, smartphones, and others. c Using sensors, mobile applications, and other tools to acquire various digital biomarker data. d Extracting and analyzing feature data such as movement, speech, eye movement, and physiological signals. e Selecting appropriate machine learning and AI algorithms to train predictive models. f Applying the models for disease classification and prediction, aiding in early detection and management.CC BY-NC-ND · PubMed Central
Fig. 17 Spatial-temporal distribution of research output. a Distribution of research years in AI model-based studies on digital biomarkers for Alzheimer’s disease diagnosis and prediction. The red dashed line represents the trend line (aTrend line:y = 0.2026x2–1.549x + 3.0714). b Regional distribution of AI model-based studies on digital biomarkers for Alzheimer’s disease diagnosis and prediction.
Fig. 17 Spatial-temporal distribution of research output. a Distribution of research years in AI model-based studies on digital biomarkers for Alzheimer’s disease diagnosis and prediction. The red dashed line represents the trend line (aTrend line:y = 0.2026x2–1.549x + 3.0714). b Regional distribution of AI model-based studies on digital biomarkers for Alzheimer’s disease diagnosis and prediction.CC BY-NC-ND · PubMed Central
Fig. 18 Distribution of AI Model Algorithms Used in Alzheimer’s Digital Biomarkers. a Overall Use of AI model algorithms. b Distribution of AI model algorithm use across different types of digital biomarkers. c Use of AI model algorithms in comparative studies between models. d Use of AI model algorithms in comparative studies of different types of digital biomarkers.
Fig. 18 Distribution of AI Model Algorithms Used in Alzheimer’s Digital Biomarkers. a Overall Use of AI model algorithms. b Distribution of AI model algorithm use across different types of digital biomarkers. c Use of AI model algorithms in comparative studies between models. d Use of AI model algorithms in comparative studies of different types of digital biomarkers.CC BY-NC-ND · PubMed Central
Fig. 19 Sunburst Chart Showing The Distribution of Specific Features Used in Various Digital Biomarker Studies. Pie segments of the same color represent the features used in research on the same type of digital biomarker. The size of the segments indicates the frequency of feature usage, with larger areas reflecting more frequent usage.
Fig. 19 Sunburst Chart Showing The Distribution of Specific Features Used in Various Digital Biomarker Studies. Pie segments of the same color represent the features used in research on the same type of digital biomarker. The size of the segments indicates the frequency of feature usage, with larger areas reflecting more frequent usage.CC BY-NC-ND · PubMed Central
Fig. 20 The Box plot of Binary Classification Performance Based on AD and MCI AI Models. Each point in the box plot represents an individual study. The horizontal line inside the box indicates the median of the data, while the whiskers represent the range of the maximum and minimum values. Data points outside the whiskers are considered outliers.
Fig. 20 The Box plot of Binary Classification Performance Based on AD and MCI AI Models. Each point in the box plot represents an individual study. The horizontal line inside the box indicates the median of the data, while the whiskers represent the range of the maximum and minimum values. Data points outside the whiskers are considered outliers.CC BY-NC-ND · PubMed Central
Fig. 21 Box plot of AI Model Performance for Binary Classification Based on AD or MCI versus HC. a AUC for AD and HC Classification, (b) ACC, (c) SEN, (d) SPE. e AUC for MCI and HC classification, (f) ACC, (g) SEN, (h) SPE.
Fig. 21 Box plot of AI Model Performance for Binary Classification Based on AD or MCI versus HC. a AUC for AD and HC Classification, (b) ACC, (c) SEN, (d) SPE. e AUC for MCI and HC classification, (f) ACC, (g) SEN, (h) SPE.CC BY-NC-ND · PubMed Central
Fig. 22 Discussion on Digital Biomarker Research for Alzheimer’s Disease. a The current multidimensional landscape of digital biomarkers in Alzheimer’s disease. b The five key dimensions of the challenges in clinical implementation of digital biomarkers. c The relationship between the main challenges, current status, and future prospects in the field of digital biomarkers for AD. The color of the circles represents the hierarchical relationships of the topics, while the arrows indicate the connections and directional relationships between all topics.
Fig. 22 Discussion on Digital Biomarker Research for Alzheimer’s Disease. a The current multidimensional landscape of digital biomarkers in Alzheimer’s disease. b The five key dimensions of the challenges in clinical implementation of digital biomarkers. c The relationship between the main challenges, current status, and future prospects in the field of digital biomarkers for AD. The color of the circles represents the hierarchical relationships of the topics, while the arrows indicate the connections and directional relationships between all topics.CC BY-NC-ND · PubMed Central
Fig. 23 Flowchart and analytical content of the bibliometrics and content analysis method used in this study. The image illustrates the three main issues addressed by the study, the analytical workflow, and the tools and methods employed.
Fig. 23 Flowchart and analytical content of the bibliometrics and content analysis method used in this study. The image illustrates the three main issues addressed by the study, the analytical workflow, and the tools and methods employed.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

1 retracted

Show all 207 references

Research works printed but not linked: 5

  1. Collins, G. S., Reitsma, J. B., Altman, D. G. & Moons, K. G. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ (Clin. Res. ed.) 350, g7594 (2015).
  2. Lazzeretti, L. & Capone, F. How proximity matters in innovation networks dynamics along the cluster evolution. A study high. Technol. Appl. cultural goods. 69, 5855–5865 (2016).
  3. Xiangmao, C., Cheng, P., Guoliang, X., Tian, H. & Gang, Z. iSleep: A Smartphone System for Unobtrusive Sleep Quality Monitoring. ACM Trans. Sens. Netw. 16, 27 (2020). (32 pp.)-27 (32 pp.).
  4. Aijing L, S. Y. & Lu M. Medical Literature Information Retrieval. Third Edition., (People’s Medical Publishing House, 2005).
  5. Garrett, S. J. An Introduction to the Mathematics of Finance: A Deterministic Approach. (Elsevier Science, 2013).

Cited by

See all 78

Metadata

Authors
Qi, W., Zhu, X., Wang, B., Shi, Y., Dong, C., Shen, S., Li, J., Zhang, K., He, Y., Zhao, M., Yao, S., Dong, Y., Shen, H., Kang, J., Lu, X., Jiang, G., Boots, L. M. M., Fu, H., Pan, L., Chen, H., Yan, Z., Xing, G., & Cao, S.
Affiliations
  • Hangzhou Normal University (CN)
  • Fujian Medical University (CN)
  • Fudan University (CN)
  • Zhejiang Chinese Medical University (CN)
  • Zhejiang Provincial People's Hospital (CN)
  • Maastricht University (NL)
  • Stony Brook University (US)
  • University of Hong Kong (HK)
  • Chinese University of Hong Kong (HK)
Corresponding author
  • Xing, Guoliang
  • Cao, Shihua
Venue
npj Digital Medicine
Published
2025-06-16
Volume
8
Issue
1
Pages
366-366
ISSN
2398-6352 (ISSN-L)
OpenAlex
W4411324838
Cited by
78
References
206
Citations by year
2026: 64 · 2025: 14
FWCI
52.48
Type
journal article
Language
en
Open access
Licence
Locations
  • npj Digital Medicine (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Research Publications (Maastricht University) (publishedVersion, cc-by-nc-nd)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
Keywords
Alzheimer's disease, digital biomarkers, artificial intelligence models, mild cognitive impairment, motor activity, neurocognitive tests, eye tracking, speech analysis, machine learning, external validation, model calibration, early diagnosis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Alzheimer's disease digital biomarkers multidimensional landscape and AI model scoping review. https://app.cochat.ai/papers/10.1038/s41746-025-01640-z
The paper
Qi, W., Zhu, X., Wang, B., Shi, Y., Dong, C., Shen, S., Li, J., Zhang, K., He, Y., Zhao, M., Yao, S., Dong, Y., Shen, H., Kang, J., Lu, X., Jiang, G., Boots, L. M. M., Fu, H., Pan, L., . . . Cao, S. (2025). Alzheimer's disease digital biomarkers multidimensional landscape and AI model scoping review. npj Digital Medicine. https://doi.org/10.1038/s41746-025-01640-z

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.