Wearable-Based Behavioural Monitoring for Dementia Care: From Dataset Development to Routine-Aware Anomaly Detection
DOI:
https://doi.org/10.4108/dtip.15188Keywords:
Wearable Sensors; Human Activity Recognition; Dataset Development; Behavioural Anomaly Detection; Synthetic Data Generation.Abstract
INTRODUCTION: Disruption of daily routines, such as irregular sleep, skipped meals or reduced mobility, is often among the earliest observable markers of functional decline in dementia, yet continuous, unobtrusive monitoring of this disruption remains difficult to build and validate at scale, mainly due to the scarcity of realistic wearable datasets and of longitudinal behavioural data.
OBJECTIVES: This paper reports two complementary contributions developed within the same wearable-monitoring project: the incremental construction of a real-world multimodal wearable dataset for human activity recognition, and a routine-aware behavioural monitoring framework that uses that dataset to train and evaluate a sequential anomaly detection model.
METHODS: A dual-device wearable setup, combining a wrist-worn inertial prototype with an external spot-check physiological monitor, was used to collect labelled activities of daily living from twelve adult participants under supervised conditions. The resulting activity database seeds a knowledge-informed synthetic daily-routine generator and trains a Bidirectional Gated Recurrent Unit (BiGRU) sequential model for context-dependent anomaly detection, evaluated on synthetic routines with injected behavioural deviations.
RESULTS: Activity-recognition baselines trained on the dataset reach an accuracy of 0.73 (LightGBM), while the anomaly detection model attains an AUROC of 0.999 and a PR-AUC of 0.989 on synthetic routines with injected behavioural deviations.
CONCLUSION: These results support the feasibility of the acquisition pipeline and of the routine-modelling approach as a foundation for future validation with longitudinal, real-world dementia cohorts.
Downloads
References
World Health Organization, “Dementia,” fact sheet, Mar. 2025. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/dementia. Accessed: Aug. 7, 2026.
[2] Alzheimer’s Disease International, “World Alzheimer Reports.” [Online]. Available: https://www.alzint.org/what-we-do/research/world-alzheimer-report/. Accessed: Aug. 7, 2026.
[3] D. Gebhard and J. I. Frank, “Everyday life and boredom of people living with dementia in residential long-term care: a merged methods study,” BMC Geriatr., vol. 24, no. 1, Dec. 2024, doi: 10.1186/s12877-024-05641-7.
[4] S.-S. Huang, “Depression among caregivers of patients with dementia: Associative factors and management approaches,” World J. Psychiatry, vol. 12, no. 1, p. 59, Jan. 2022, doi: 10.5498/WJP.V12.I1.59.
[5] I. C. Rocha et al., “Monitoring Wearable Devices for Elderly People with Dementia: A Review,” Aug. 01, 2024, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/designs8040075.
[6] M. Carvalho et al., “Powered Wearable Technologies for Dementia Care: Evaluating Activity Recognition Models and Dataset Challenges,” INSTICC, Feb. 2025, pp. 995–1006. doi: 10.5220/0013396600003911.
[7] A. C. Cote, R. J. Phelps, N. S. Kabiri, J. S. Bhangu, and K. Thomas, “Evaluation of Wearable Technology in Dementia: A Systematic Review and Meta-Analysis,” Jan. 11, 2021, Frontiers Media S.A. doi: 10.3389/fmed.2020.501104.
[8] J. Kim, A. S. Campbell, B. E. F. de Ávila, and J. Wang, “Wearable biosensors for healthcare monitoring,” Apr. 01, 2019, Nature Publishing Group. doi: 10.1038/s41587-019-0045-y.
[9] CarePredict Inc., “CarePredict @Home.” [Online]. Available: https://www.carepredict.com/go. Accessed: Aug. 7, 2026.
[10] Feel Tranquil, “Tranquil GPS Watch for Elderly / Seniors & Dementia Patients.” [Online]. Available: https://feeltranquil.com/gps-tracking-watch-elderly-seniors/. Accessed: Aug. 7, 2026.
[11] UnaliWear Inc., “The Kanega Watch®: Medical Alert Watch with Fall Detection.” [Online]. Available: https://www.unaliwear.com/product/kanega-watch/. Accessed: Aug. 7, 2026.
[12] A. Pappadà, R. Chattat, I. Chirico, M. Valente, and G. Ottoboni, “Assistive Technologies in Dementia Care: An Updated Analysis of the Literature,” Mar. 24, 2021, Frontiers Media S.A. doi: 10.3389/fpsyg.2021.644587.
[13] A. Reiss and D. Stricker, “PAMAP2 Physical Activity Monitoring,” UCI Machine Learning Repository, 2012, doi: 10.24432/C5NW2H. [Dataset]. [Online]. Available: https://archive.ics.uci.edu/dataset/231/pamap2+physical+activity+monitoring. Accessed: Sep. 1, 2025.
[14] D. Roggen et al., “OPPORTUNITY Activity Recognition,” UCI Machine Learning Repository, 2010, doi: 10.24432/C5M027. [Dataset]. [Online]. Available: https://archive.ics.uci.edu/dataset/226/opportunity+activity+recognition. Accessed: Sep. 1, 2025.
[15] J. Varghese, A. Brenner, L. Plagwitz, C. van Alen, M. Fujarski, and T. Warnecke, “PADS - Parkinsons Disease Smartwatch dataset,” version 1.0.0, PhysioNet, 2024, doi: 10.13026/m0w9-zx22. [Dataset]. [Online]. Available: https://physionet.org/content/parkinsons-disease-smartwatch/1.0.0/. Accessed: Sep. 1, 2025.
[16] kaggler2412, “Dementia Patient Health and Prescriptions ML Dataset,” Kaggle, 2024. [Dataset]. [Online]. Available: https://www.kaggle.com/datasets/kaggler2412/dementia-patient-health-and-prescriptions-dataset. Accessed: Sep. 1, 2025.
[17] M. Smuck, C. A. Odonkor, J. K. Wilt, N. Schmidt, and M. A. Swiernik, “The emerging clinical role of wearables: factors for successful implementation in healthcare,” Dec. 01, 2021, Nature Research. doi: 10.1038/s41746-021-00418-3.
[18] L. Cheng, Y. Guan, K. Zhu, and Y. Li, “Recognition of human activities using machine learning methods with wearable sensors,” Sep. 2017, pp. 1–7. doi: 10.1109/CCWC.2017.7868369.
[19] Ó. D. Lara and M. A. Labrador, “A survey on human activity recognition using wearable sensors,” IEEE Communications Surveys and Tutorials, vol. 15, no. 3, pp. 1192–1209, 2013, doi: 10.1109/SURV.2012.110112.00192.
[20] Ç. Erdaş and S. Güney, “Human Activity Recognition by Using Different Deep Learning Approaches for Wearable Sensors,” Neural Process. Lett., vol. 53, Sep. 2021, doi: 10.1007/s11063-021-10448-3.
[21] I. U. Khan, S. Afzal, and J. W. Lee, “Human activity recognition via hybrid deep learning based model,” Sensors, vol. 22, no. 1, Jan. 2022, doi: 10.3390/s22010323.
[22] M. Karas, J. Urbanek, C. Crainiceanu, J. Harezlak, and W. Fadel, “Labeled raw accelerometry data captured during walking, stair climbing and driving,” version 1.0.0, PhysioNet, 2021, doi: 10.13026/51h0-a262. [Dataset]. [Online]. Available: https://physionet.org/content/accelerometry-walk-climb-drive/1.0.0/. Accessed: Sep. 1, 2025.
[23] E. Grimaldi, D. Vigneri, G. Di Poce, and N. Grieco, “Falls vs Normal Activities,” Kaggle, 2023. [Dataset]. [Online]. Available: https://www.kaggle.com/datasets/enricogrimaldi/falls-vs-normal-activities. Accessed: Sep. 1, 2025.
[24] J. Midroni et al., “Prediction of Incident Alzheimer’s Disease in Older Community‐Dwelling Adults using Wrist Accelerometry and Deep Learning,” Alzheimer’s & Dementia, vol. 20, no. S2, Dec. 2024, doi: 10.1002/ALZ.087666.
[25] B. Leite, A. Abdalrahman, J. Castro, J. Frade, J. Moreira, and C. Soares, “Novelty detection in physical activity,” in ICAART 2021 - Proceedings of the 13th International Conference on Agents and Artificial Intelligence, SciTePress, 2021, pp. 859–865. doi: 10.5220/0010254908590865.
[26] J. Xu, H. Wu, J. Wang, and M. Long, “Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy,” in International Conference on Learning Representations (ICLR), 2022. [Online]. Available: https://openreview.net/forum?id=LzQQ89U1qm_. Accessed: Aug. 7, 2026.
[27] W. Zhang and C. Luo, “Decomposition-based multi-scale transformer framework for time series anomaly detection,” 2025. [Online]. Available: https://github.com/shaieesss/TransDe. Accessed: Aug. 7, 2026.
[28] M. Amiribesheli and H. Bouchachia, “A tailored smart home for dementia care,” J. Ambient Intell. Humaniz. Comput., vol. 9, no. 6, pp. 1755–1782, Nov. 2018, doi: 10.1007/s12652-017-0645-7.
[29] A. Iaboni et al., “Wearable multimodal sensors for the detection of behavioral and psychological symptoms of dementia using personalized machine learning models,” Alzheimer’s and Dementia: Diagnosis, Assessment and Disease Monitoring, vol. 14, no. 1, 2022, doi: 10.1002/dad2.12305.
[30] M. Loni, F. Poursalim, M. Asadi, and A. Gharehbaghi, “A review on generative AI models for synthetic medical text, time series, and longitudinal data Check for updates,” 2025, doi: 10.1038/s41746-024-01409-w.
[31] M. Miletic and M. Sariyar, “Synthetic data generation methods for longitudinal and time series health data: a systematic review,” BMC Med. Inform. Decis. Mak., 2026, doi: 10.1186/s12911-025-03326-8.
[32] M. I. Rahman, M. R. Hossain, S. M. Sayem, and F. Bin Emdad, “Exploring the role of synthetic data in the future of AI in healthcare: A scoping review of frameworks, challenges, and implications,” Mar. 01, 2026, Elsevier B.V. doi: 10.1016/j.ibmed.2025.100342.
[33] H. Murtaza, M. Ahmed, N. F. Khan, G. Murtaza, S. Zafar, and A. Bano, “Synthetic data generation: State of the art in health care domain,” May 01, 2023, Elsevier Ireland Ltd. doi: 10.1016/j.cosrev.2023.100546.
[34] M. Carvalho et al., “Early stage development of a wearable-based dataset for human activity recognition,” in Innovations in Mechatronics Engineering V, J. Machado, J. Trojanowska, F. Soares, P. Rea, S. Butdee, and B. Gramescu, Eds. Cham, Switzerland: Springer, 2026, Lecture Notes in Mechanical Engineering. [Online]. Available: https://link.springer.com/book/9783032293749.
[35] S. Enshaeifar et al., “Health management and pattern analysis of daily living activities of people with dementia using in-home sensors and machine learning techniques,” PLoS One, vol. 13, no. 5, May 2018, doi: 10.1371/journal.pone.0195605.
[36] D. Arifoglu and A. Bouchachia, “Activity Recognition and Abnormal Behaviour Detection with Recurrent Neural Networks,” in Procedia Computer Science, Elsevier B.V., 2017, pp. 86–93. doi: 10.1016/j.procs.2017.06.121.
[37] D. Arifoglu and A. Bouchachia, “Detection of abnormal behaviour for dementia sufferers using Convolutional Neural Networks,” Artif. Intell. Med., vol. 94, pp. 88–95, Mar. 2019, doi: 10.1016/J.ARTMED.2019.01.005.
[38] A. Lotfi, C. Langensiepen, S. M. Mahmoud, and M. J. Akhlaghinia, “Smart homes for the elderly dementia sufferers: identification and prediction of abnormal behaviour,” J. Ambient Intell. Humaniz. Comput., vol. 3, no. 3, pp. 205–218, Sep. 2012, doi: 10.1007/s12652-010-0043-x.
[39] M. Arantes et al., “Towards the Development of a Wearable Device Enhancing Inclusivity and Monitoring for Dementia Patients,” in Lecture Notes in Mechanical Engineering, Springer Science and Business Media Deutschland GmbH, 2025, pp. 360–370. doi: 10.1007/978-3-031-94223-5_32.
[40] J. Soares et al., “Development of an In-Ear Wearable Device for Continuous Monitoring of Elderly Patients with Dementia: A Preliminary Approach,” in Proceedings of the 19th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2026), vol. 2: BIODEVICES, SciTePress, 2026, pp. 206–213, ISBN 978-989-758-802-0. [Online]. Available: https://www.scitepress.org/PublishedPapers/2026/146520/. Accessed: Aug. 7, 2026.
[41] Shenzhen Creative Industry Co., Ltd., “Spot-Check Monitor User Instruction Manual,” ver. 1.3, Dec. 1, 2016. [Online]. Available: https://es.creative-sz.com/uploads/file/pc-200-manual.pdf. Accessed: Aug. 7, 2026.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Mariana Carvalho, Inês Rocha, Marcelo Arantes, Ana Freitas, José Soares, Vanessa Baeta, Demétrio Matos, Pedro Morais, Vìtor Carvalho
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
