A review of AI Technologies for Wearable Devices
Open accessAbstract
With the popularity of wearable devices, we can collect various data to support a series of innovative applications. The complex and massive data requires stronger data processing technologies. In recent years, artificial intelligence technology has been used to process this rich but complex data. In this paper, we summarize the research of AI technologies for wearable devices, from the aspects of types of wearable devices, collected data, models, and applications. We find that artificial intelligence technology has not only made a breakthrough in performance over traditional methods, but also creates a series of new applications. For the future research directions, we also point out some problems, e.g., the sensor data measurement and classification are not accurate enough, which would inspires the following research to investigate further.
AI snapshot
Written from the paper’s full text.
TL;DR
This review summarizes AI technologies applied to wearable devices, covering device types, collected data, models, and applications. It concludes that AI has enabled new wearable applications but identifies ongoing challenges in data accuracy, noise, individual differences, and energy constraints.
Key findings
- Wearable devices reviewed include smart phones, smart watches/wristbands, smart glasses, smart clothes/socks, smart shoes, and smart earphones; collected data span sports, physiological, environmental, and communication data.
- The paper highlights three model families: machine learning/ensemble learning, convolutional neural networks, and long short-term memory networks.
- Applications summarized include monitoring of motion state, recognition of sports behavior, health care, pressure test, and sports monitoring.
- The paper reports no quantitative performance results; it states that AI has made breakthroughs over traditional methods but does not provide numerical comparisons.
Methods
- Narrative review: the author searched for relevant papers from IEEE, ACM, Springer, and Elsevier via academic search engines such as Google Scholar, using keywords such as artificial intelligence and wearable devices.
- The review organizes prior work by wearable device types, data types, AI models, and applications, and then discusses future research directions.
- No systematic search protocol, inclusion/exclusion criteria, sample size, or quantitative synthesis is reported.
Limitations
- Stated by the paper: sensor data measurement and classification are not accurate enough; laboratory-collected data may not generalize to real data; test data are imbalanced; collected data often contain noise; algorithms depend on device position and placement mode; wearing many devices causes discomfort and consumes electricity; video and environmental sensors have privacy, location, and noise constraints; individual behavioral differences reduce classification accuracy; and wearable devices have limited battery and storage resources.
- Evident from the text: as a narrative review, it does not provide a systematic search strategy, quality appraisal, or quantitative synthesis, and it reports no performance metrics from the reviewed studies.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 3 cited sources checked · 0 supported · 0 flagged
- Findingnone given
The data used in the experiment are collected in the laboratory.
“The data used in the experiment are collected in the laboratory, indicating that when it comes to real data, the effect may be discounted to some extent.”
Quote in the paperSubstantial claim - Findingnone given
There is an imbalance in the test data.
“There is an imbalance in the test data, and the percentage of each exercise in the training data is different.”
Quote in the paperSubstantial claim - Interpretationreasoning
Artificial intelligence technology has created a series of new applications for wearable devices.
“We find that artificial intelligence technology has not only made a breakthrough in performance over traditional methods, but also creates a series of new applications.”
- Also stated in: § 6. Conclusion and Future Directions · plain
“In this paper, we find that AI technologies have been successfully applied for wearable devices and inspired a various types of applications.”
Quote in the paperSubstantial claim - Also stated in: § 6. Conclusion and Future Directions · plain
- Interpretationnone given
AI technologies have been successfully applied for wearable devices.
“In this paper, we find that AI technologies have been successfully applied for wearable devices and inspired a various types of applications.”
Quote in the paperSubstantial claim - Interpretationnone given
Artificial intelligence technology has made a breakthrough in performance over traditional methods for wearable devices.
“We find that artificial intelligence technology has not only made a breakthrough in performance over traditional methods, but also creates a series of new applications.”
Quote in the paperNot a substantial claim
- Interpretationreasoning
Wearing a large number of devices in different parts of the body may cause discomfort.
“Effective activity identification of wearable devices requires users to wear a large number of devices in different parts of the body, which may cause discomfort and consumes lots of electricity.”
Quote in the paperSubstantial claim - Interpretationreasoning
When it comes to real data, the effect may be discounted to some extent.
“The data used in the experiment are collected in the laboratory, indicating that when it comes to real data, the effect may be discounted to some extent.”
Quote in the paperNot a substantial claim - Interpretationnone given
The sensor based on video captures non-target information.
“The sensor based on video infringes users' privacy, restricts users to a specific location and captures non-target information.”
Quote in the paperSubstantial claim - Interpretationreasoning
Using unbalanced data for training may bring negative impact on the results.
“So using the unbalanced data for training may bring negative impact on the results.”
Quote in the paperSubstantial claim - Interpretationreasoning
The sensor based on video restricts users to a specific location.
“The sensor based on video infringes users' privacy, restricts users to a specific location and captures non-target information.”
Quote in the paperSubstantial claim - Interpretationreasoning
Video and environmental sensors are not suitable for normal activities.
“Video and environmental sensors work under fixed conditions and are not suitable for normal activities.”
Quote in the paperSubstantial claim - Interpretationnone given
Video and environmental sensors work under fixed conditions.
“Video and environmental sensors work under fixed conditions and are not suitable for normal activities.”
Quote in the paperSubstantial claim - Interpretationreasoning
Current methodologies for AI in wearable devices still face many challenges.
“However, current methodologies still face many challenges, which need further exploration.”
Quote in the paperSubstantial claim
- Classified as probably not a substantial claim. 10
- May combine two claims. 1
Classified as probably not a substantial claim.
Findingnone givenThe percentage of each exercise in the training data is different.
“There is an imbalance in the test data, and the percentage of each exercise in the training data is different.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
Interpretationnone givenSensor data measurement and classification for wearable devices are not accurate enough.
“For the future research directions, we also point out some problems, e.g., the sensor data measurement and classification are not accurate enough, which would inspires the following research to investigate further.”
- Also stated in: § 6. Conclusion and Future Directions · plain
“(1) Inaccurate sensor data measurement”
Quote in the paperSubstantial claim- Also stated in: § 6. Conclusion and Future Directions · plain
Classified as probably not a substantial claim.
Interpretationnone givenDue to limitations of wearable devices and interference of external environment, the collected data often contain a lot of noise.
“Due to the limitations of wearable devices and the interference of external environment, the collected data often contain a lot of noise.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenEffective activity identification of wearable devices requires users to wear a large number of devices in different parts of the body.
“Effective activity identification of wearable devices requires users to wear a large number of devices in different parts of the body, which may cause discomfort and consumes lots of electricity.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
Interpretationnone givenThe sensor based on video infringes users' privacy.
“The sensor based on video infringes users' privacy, restricts users to a specific location and captures non-target information.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationDeep learning methods have achieved performance far beyond traditional methods in fields such as image recognition, audio processing, and traffic prediction.
“Especially in recent years, with the rise of deep learning [1] methods represented by a series of deep neural networks, more and more artificial intelligence technologies begin to play an important role in various fields, such as image recognition [2], audio processing [3], traffic prediction [4], and have achieved positive performance far beyond traditional methods.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [1] LeCun, 2015, Deep learning[J], nature source
Not checked in the free run.
- cites [2] He, 2016, Deep residual learning for image recognition[C]
Not checked in the free run.
- cites [3] Yue-Hei, 2015, Beyond short snippets: Deep networks for video classification[C]
Not checked in the free run.
Probably not a substantial claim: may not pass S4 (based).
Interpretationnone givenWearing a large number of devices in different parts of the body consumes lots of electricity.
“Effective activity identification of wearable devices requires users to wear a large number of devices in different parts of the body, which may cause discomfort and consumes lots of electricity.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenThe difference of individual behaviors leads to the low accuracy of traditional model identification.
“The difference of individual behaviors leads to the low accuracy of traditional model identification.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenExisting algorithms are closely related to the position and mode to place the device.
“Wearable devices are used casually in daily life, and existing algorithms are closely related to the position and mode to place the device.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Interpretationnone givenEnvironmental noise will affect the performance of environmental sensor.
“Meanwhile, environmental noise will affect the performance of environmental sensor.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
Traditional data processing methods are unable to meet the processing requirements of new wearable device applications.
“Traditional methods of data processing are unable to meet the processing requirements of new applications.”
The paper relies on this without support.
Wearable device data are complex and massive and require stronger data processing technologies.
“The complex and massive data requires stronger data processing technologies.”
The paper relies on this without support.
Wearable devices have very limited battery and storage resources.
“The battery and storage resources of wearable devices are very limited”
The paper relies on this without support.
Ensemble learning can often achieve better results than single models.
“Through this combination, ensemble learning can often achieve better results than single models.”
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?
Only a few studies have focused on the relationship between biofluid signals and human health parameters, such as cortisol sensing and stress, glucose sensing and diabetes, or physiological data related to Parkinson's disease.
“Only a few studies have focused on this, such as one recent work on cortisol sensing and its relation to human stress levels [129] and glucose sensing and its relation to diabetes levels in patients [252], or physiological data related to Parkinson’s disease [253].”
From The emergence of AI-based wearable sensors for digital health technology: a review (2023) · cites this paper as [253]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Audio-processing example cites a video-classification paper
MinorThe introduction cites reference [3] as an example of audio processing, but the reference is titled as a study of video classification.
“such as image recognition [2], audio processing [3], traffic prediction [4]”
“3. Yue-Hei Ng J, Hausknecht M, Vijayanarasimhan S, et al. Beyond short snippets: Deep networks for video classification[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 4694-4702.”
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126 · see citing works
Deep learning for healthcare: review, opportunities and challenges
Miotto, R., Wang, F., Wang, S., Jiang, X., & Dudley, J. T. (2018) · Briefings in Bioinformatics · doi:10.1093/bib/bbx044 · Cited by 3,076 · see citing works
Beyond short snippets: Deep networks for video classification
Ng, J. Y.-H., Hausknecht, M., Vijayanarasimhan, S., Vinyals, O., Monga, R., & Toderici, G. D. (2015) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2015.7299101 · Cited by 2,121 · see citing works
Deep learning for sensor-based activity recognition: A survey
Wang, J., Chen, Y., Hao, S., Peng, X., & Hu, L. (2019) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.02.010 · Cited by 2,014 · see citing works
Nweke, H. F., Teh, Y. W., Al-garadi, M. A., & Alo, U. R. (2018) · Expert Systems with Applications · doi:10.1016/j.eswa.2018.03.056 · Cited by 930 · see citing works
Stress Recognition Using Wearable Sensors and Mobile Phones
Sano, A., & Picard, R. W. (2013) · 2013 Humaine Association Conference on Affective Computing and Intelligent Interaction · doi:10.1109/acii.2013.117 · Cited by 523 · see citing works
A deep learning approach to on-node sensor data analytics for mobile or wearable devices
Ravi, D., Wong, C., Lo, B., & Yang, G.-Z. (2017) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2016.2633287 · Cited by 439 · see citing works
Detecting Falls with Wearable Sensors Using Machine Learning Techniques
Özdemir, A. T., & Barshan, B. (2014) · Sensors · doi:10.3390/s140610691 · Cited by 376 · see citing works
Steven Eyobu, O., & Han, D. S. (2018) · Sensors · doi:10.3390/s18092892 · Cited by 259 · see citing works
Kubota, K. J., Chen, J. A., & Little, M. A. (2016) · Movement Disorders · doi:10.1002/mds.26693 · Cited by 210 · see citing works
Geospatial data to images: a deep-learning framework for traffic forecasting
Jiang, W., & Zhang, L. (2019) · Tsinghua Science and Technology · doi:10.26599/tst.2018.9010033 · Cited by 176 · see citing works
Towards multimodal deep learning for activity recognition on mobile devices
Radu, V., Lane, N. D., Bhattacharya, S., Mascolo, C., Marina, M. K., & Kawsar, F. (2016) · Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct · doi:10.1145/2968219.2971461 · Cited by 147 · see citing works
Human Activity Recognition from Body Sensor Data using Deep Learning
Hassan, M. M., Huda, S., Uddin, M. Z., Almogren, A., & Alrubaian, M. (2018) · Journal of Medical Systems · doi:10.1007/s10916-018-0948-z · Cited by 113 · see citing works
Iss2Image: A Novel Signal-Encoding Technique for CNN-Based Human Activity Recognition
Hur, T., Bang, J., Huynh‐The, T., Lee, J., Kim, J., & Lee, S. (2018) · Sensors · doi:10.3390/s18113910 · Cited by 74 · see citing works
A Wearable Activity Recognition Device Using Air-Pressure and IMU Sensors
Yang, D., Huang, J., Tu, X., Ding, G., Shen, T., & Xiao, X. (2018) · IEEE Access · doi:10.1109/access.2018.2890004 · Cited by 46 · see citing works
Dargazany, A. R., Stegagno, P., & Mankodiya, K. (2018) · Mobile Information Systems · doi:10.1155/2018/8125126 · Cited by 45 · see citing works
Activity-aware essential tremor evaluation using deep learning method based on acceleration data
Zheng, X., Vieira, A., Marcos, S. L., Aladro, Y., & Ordieres-Meré, J. B. (2018) · Parkinsonism & Related Disorders · doi:10.1016/j.parkreldis.2018.08.001 · Cited by 38 · see citing works
Movement and Gesture Recognition Using Deep Learning and Wearable-sensor Technology
Xie, B., Li, B., & Harland, A. R. (2018) · doi:10.1145/3268866.3268890 · Cited by 33 · see citing works
Prawiro, E. A. P. J., Chou, N., Lee, M.-W., et al. (2019) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2018.2880829 · Cited by 21 · see citing works
Noury, N., Dittmar, A., Corroy, C., et al. (2004) · doi:10.1109/health.2004.1324507 · Cited by 20 · see citing works
Wearable Sensor Data Based Human Activity Recognition using Machine Learning: A new approachPreprint
Nguyen, H. D., Tran, K. P., Zeng, X., Koehl, L., & Tartare, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.03809 · Cited by 17 · see citing works
Development of piezoelectric vibration energy harvesters for battery-less smart shoes
Katsumura, H., Konishi, T., Okumura, H., et al. (2018) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1052/1/012060 · Cited by 10 · see citing works
Deep learning approaches for human activity recognition using wearable technology
Janković, M. M., Savić, A. M., Novicic, M. M., & Popović, M. (2018) · Medicinski podmladak · doi:10.5937/mp69-18039 · Cited by 6 · see citing works
Wearable Devices for Human Activity Recognition and User Detection
Capobianco, G., Giacomo, U. D., Martinelli, F., Mercaldo, F., & Santone, A. (2019) · Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP) · doi:10.1109/empdp.2019.8671636 · Cited by 6 · see citing works
Flagged first, then by citations
Research works printed but not linked: 4
- Zheng, 2018, Review of human activity recognition based on wearable sensors[J], Journal of Computer Applications, 5, 1223
- Hu, 2018, Fall detection algorithms based on wearable device: a review[J], Journal of Zhejiang University (Engineering Science), 52, 1717
- Moya Rueda, 2018, 5, 26
- Chen, 2018, On deep-learning-model-based sensor activity recognition[J], Journal of Zhejiang University of Technology, 46, 375
Cited by
Sorted by citations
The emergence of AI-based wearable sensors for digital health technology: a review
Shajari, S., Kuruvinashetti, K., Komeili, A., & Sundararaj, U. (2023) · Sensors · doi:10.3390/s23239498 · Cited by 535 · see citing works
- Authors
- Jin, C. Y.
- Affiliations
- Northwestern Polytechnical University (CN)
- Corresponding author
- Jin, Chun Yu
- Venue
- IOP Conference Series: Materials Science and Engineering
- Publisher
- IOP Publishing
- Published
- 2019-11-01
- Volume
- 688
- Issue
- 4
- Pages
- 044072
- ISSN
- 1757-8981 (ISSN-L)
- 1757-899X
- OpenAlex
- W2994593995
- Cited by
- 32
- References
- 25
- Citations by year
- 2026: 5 · 2025: 5 · 2024: 7 · 2023: 5 · 2022: 6 · 2021: 4
- FWCI
- 0.81
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- IOP Conference Series Materials Science and Engineering (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Information Systems › Internet of Things and AI
- Physical Sciences › Engineering › Biomedical Engineering › Advanced Sensor and Energy Harvesting Materials
- Keywords
- wearable devices, artificial intelligence, wearable sensors, sensor data classification
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.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.
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
