ABSTRACT
Self-tracking using commodity wearables such as smartwatches can help older adults reduce sedentary behaviors and engage in physical activity. However, activity recognition applications that are typically deployed in these wearables tend to be trained on datasets that best represent younger adults. We explore how our activity recognition model, a hybrid of long short-term memory and convolutional layers, pre-trained on smartwatch data from younger adults, performs on older adult data. We report results on week-long data from two older adults collected in a preliminary study in the wild with ground-truth annotations based on activPAL, a thigh-worn sensor. We find that activity recognition for older adults remains challenging even when comparing our model’s performance to state of the art deployed models such as the Google Activity Recognition API. More so, we show that models trained on younger adults tend to perform worse on older adults.
- [n.d.]. Google Activity Recognition API. https://developers.google.com/location-context/activity-recognition/Google Scholar
- 2001. PAL Technologies. https://www.palt.com/Google Scholar
- Rebecca Adaimi and Edison Thomaz. 2019. Leveraging active learning and conditional mutual information to minimize data annotation in human activity recognition. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 3, 3 (2019), 1–23.Google ScholarDigital Library
- Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra, Jorge Luis Reyes-Ortiz, 2013. A public domain dataset for human activity recognition using smartphones.. In Esann, Vol. 3. 3.Google Scholar
- Centers for Disease Control and Prevention. 2019. Lack of Physical Activity. https://www.cdc.gov/chronicdisease/resources/publications/factsheets/physical-activity.htmGoogle Scholar
- Rohan Chandra, Uttaran Bhattacharya, Christian Roncal, Aniket Bera, and Dinesh Manocha. 2019. RobustTP: End-to-End Trajectory Prediction for Heterogeneous Road-Agents in Dense Traffic with Noisy Sensor Inputs. In ACM Computer Science in Cars Symposium(Kaiserslautern, Germany) (CSCS ’19). Association for Computing Machinery, New York, NY, USA, Article 2, 9 pages. https://doi.org/10.1145/3359999.3360495Google ScholarDigital Library
- Yuwen Chen, Kunhua Zhong, Ju Zhang, Qilong Sun, and Xueliang Zhao. 2016. LSTM networks for mobile human activity recognition. In 2016 International conference on artificial intelligence: technologies and applications. Atlantis Press, 50–53.Google ScholarCross Ref
- Eun Kyoung Choe, Saeed Abdullah, Mashfiqui Rabbi, Edison Thomaz, Daniel A Epstein, Felicia Cordeiro, Matthew Kay, Gregory D Abowd, Tanzeem Choudhury, James Fogarty, 2017. Semi-automated tracking: a balanced approach for self-monitoring applications. IEEE Pervasive Computing 16, 1 (2017), 74–84.Google ScholarDigital Library
- Jane Chung, Blaine Reeder, Amanda Lazar, Jonathan Joe, George Demiris, and Hilaire J Thompson. 2014. Exploring an informed decision-making framework using in-home sensors: older adults’ perceptions. Journal of Innovation in Health Informatics 21, 2 (2014), 73–77.Google ScholarCross Ref
- David E Conroy, Jaclyn P Maher, Steriani Elavsky, Amanda L Hyde, and Shawna E Doerksen. 2013. Sedentary behavior as a daily process regulated by habits and intentions.Health Psychology 32, 11 (2013), 1149.Google Scholar
- Federico Cruciani, Anastasios Vafeiadis, Chris Nugent, Ian Cleland, Paul McCullagh, Konstantinos Votis, Dimitrios Giakoumis, Dimitrios Tzovaras, Liming Chen, and Raouf Hamzaoui. 2020. Feature learning for human activity recognition using convolutional neural networks. CCF Transactions on Pervasive Computing and Interaction 2, 1(2020), 18–32.Google ScholarCross Ref
- Utkarsh Dwivedi, Jaina Gandhi, Raj Parikh, Merijke Coenraad, Elizabeth Bonsignore, and Hernisa Kacorri. 2021. Exploring Machine Teaching with Children. In 2021 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC). Ieee.Google Scholar
- John Elflein. 2019. Percentage of people that used technology to track their fitness by age 2016. https://www.statista.com/statistics/742448/global-fitness-tracking-and-technology-by-age/Google Scholar
- José María Faílde Garrido, María Dolores Dapía Conde, Miguel Ángel Vázquez Vázquez, and Lucía Vázquez Rodríguez. 2021. Are Older Adults’ Leisure Patterns Consistent with the Active Ageing Model? The Influence of the Ageist Stereotypes. Social Policy and Society(2021), 1–22. https://doi.org/10.1017/S1474746421000221Google ScholarCross Ref
- Chelsea Finn, Pieter Abbeel, and Sergey Levine. 2017. Model-agnostic meta-learning for fast adaptation of deep networks. In International Conference on Machine Learning. PMLR, 1126–1135.Google Scholar
- Kathrin Gerling, Mo Ray, Vero Vanden Abeele, and Adam B Evans. 2020. Critical reflections on technology to support physical activity among older adults: An exploration of leading HCI venues. ACM Transactions on Accessible Computing (TACCESS) 13, 1 (2020), 1–23.Google ScholarDigital Library
- Jonggi Hong, Kyungjun Lee, June Xu, and Hernisa Kacorri. 2020. Crowdsourcing the Perception of Machine Teaching. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI ’20). Association for Computing Machinery, New York, NY, USA, 1–14. https://doi.org/10.1145/3313831.3376428Google ScholarDigital Library
- Hernisa Kacorri. 2017. Teachable machines for accessibility. ACM SIGACCESS Accessibility and Computing119 (2017), 10–18.Google Scholar
- Hernisa Kacorri, Utkarsh Dwivedi, and Rie Kamikubo. 2020. Data Sharing in Wellness, Accessibility, and Aging. NeurIPS 2020 Workshop on Dataset Curation and Security (2020).Google Scholar
- Hernisa Kacorri, Kris M. Kitani, Jeffrey P. Bigham, and Chieko Asakawa. 2017. People with Visual Impairment Training Personal Object Recognizers: Feasibility and Challenges. Association for Computing Machinery, New York, NY, USA, 5839–5849. https://doi.org/10.1145/3025453.3025899Google ScholarDigital Library
- Rie Kamikubo, Utkarsh Dwivedi, and Hernisa Kacorri. 2021. Sharing Practices for Datasets Related to Accessibility and Aging. The23rd International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS ’21)(2021). https://doi.org/10.1145/3441852.34712081Google ScholarDigital Library
- Young-Ho Kim, Bongshin Lee, Arjun Srinivasan, and Eun Kyoung Choe. 2021. Data@Hand: Fostering Visual Exploration of Personal Data On Smartphones Leveraging Speech and Touch Interaction. Association for Computing Machinery, New York, NY, USA. https://doi-org.proxy-um.researchport.umd.edu/10.1145/3411764.3445421Google Scholar
- William J Korotitsch and Rosemery O Nelson-Gray. 1999. An overview of self-monitoring research in assessment and treatment.Psychological Assessment 11, 4 (1999), 415.Google Scholar
- Jennifer R Kwapisz, Gary M Weiss, and Samuel A Moore. 2011. Activity recognition using cell phone accelerometers. ACM SigKDD Explorations Newsletter 12, 2 (2011), 74–82.Google ScholarDigital Library
- Matthew A Ladwig, Christopher N Sciamanna, Kayla Rutt, Joshua Blaker, Kalen Kearcher, Brandon J Auer, Liza S Rovniak, David E Conroy, Jinger S Gottschall, Matthew L Silvis, 2021. Adult outdoor group sport play during a pandemic: Feasibility, acceptability, and program adherence results from a study of modifications to mitigate COVID-19 risk. Preventive Medicine Reports(2021), 101476.Google Scholar
- Kyungjun Lee, Jonggi Hong, Simone Pimento, Ebrima Jarjue, and Hernisa Kacorri. 2019. Revisiting Blind Photography in the Context of Teachable Object Recognizers. In The 21st International ACM SIGACCESS Conference on Computers and Accessibility(Pittsburgh, PA, USA) (ASSETS ’19). Association for Computing Machinery, New York, NY, USA, 83–95. https://doi.org/10.1145/3308561.3353799Google ScholarDigital Library
- Chenglin Li, Di Niu, Bei Jiang, Xiao Zuo, and Jianming Yang. 2021. Meta-HAR: Federated Representation Learning for Human Activity Recognition. In Proceedings of the Web Conference 2021 (Ljubljana, Slovenia) (WWW ’21). Association for Computing Machinery, New York, NY, USA, 912–922. https://doi.org/10.1145/3442381.3450006Google ScholarDigital Library
- Jaclyn P Maher, Martin J Sliwinski, and David E Conroy. 2017. Feasibility and preliminary efficacy of an intervention to reduce older adults’ sedentary behavior. Translational behavioral medicine 7, 1 (2017), 52–61.Google Scholar
- Ronald Mutegeki and Dong Seog Han. 2020. A CNN-LSTM Approach to Human Activity Recognition. In 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). 362–366. https://doi.org/10.1109/ICAIIC48513.2020.9065078Google ScholarCross Ref
- Joon Sung Park, Michael S Bernstein, Robin N Brewer, Ece Kamar, and Meredith Ringel Morris. 2021. Understanding the Representation and Representativeness of Age in AI Data Sets. arXiv preprint arXiv:2103.09058(2021).Google Scholar
- Christy Pettey. 2018. Wearables Hold the Key to Connected Health Monitoring. https://www.gartner.com/smarterwithgartner/wearables-hold-the-key-to-connected-health-monitoring/Google Scholar
- Daniel Roggen, Alberto Calatroni, Mirco Rossi, Thomas Holleczek, Kilian Förster, Gerhard Tröster, Paul Lukowicz, David Bannach, Gerald Pirkl, Alois Ferscha, 2010. Collecting complex activity datasets in highly rich networked sensor environments. In 2010 Seventh international conference on networked sensing systems (INSS). IEEE, 233–240.Google ScholarCross Ref
- Hamilton Roschel, Guilherme G Artioli, and Bruno Gualano. 2020. Risk of increased physical inactivity during COVID-19 outbreak in older people: a call for actions.Google Scholar
- Allan Stisen, Henrik Blunck, Sourav Bhattacharya, Thor Siiger Prentow, Mikkel Baun Kjærgaard, Anind Dey, Tobias Sonne, and Mads Møller Jensen. 2015. Smart devices are different: Assessing and mitigatingmobile sensing heterogeneities for activity recognition. In Proceedings of the 13th ACM conference on embedded networked sensor systems. 127–140.Google ScholarDigital Library
- Pratik Tarafdar and Indranil Bose. 2021. Recognition of human activities for wellness management using a smartphone and a smartwatch: a boosting approach. Decision Support Systems 140 (2021), 113426.Google ScholarCross Ref
- Yonatan Vaizman, Katherine Ellis, and Gert Lanckriet. 2017. Recognizing Detailed Human Context in the Wild from Smartphones and Smartwatches. IEEE Pervasive Computing 16, 4 (2017), 62–74. https://doi.org/10.1109/MPRV.2017.3971131Google ScholarCross Ref
- Yonatan Vaizman, Nadir Weibel, and Gert Lanckriet. 2018. Context recognition in-the-wild: Unified model for multi-modal sensors and multi-label classification. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 1, 4 (2018), 1–22.Google ScholarDigital Library
- G Weiss. 2019. WISDM smartphone and smartwatch activity and biometrics dataset data set.Google Scholar
- Gary M Weiss, Kenichi Yoneda, and Thaier Hayajneh. 2019. Smartphone and smartwatch-based biometrics using activities of daily living. IEEE Access 7(2019), 133190–133202.Google ScholarCross Ref
- Christopher K Wong, Helena M Mentis, and Ravi Kuber. 2018. The bit doesn’t fit: Evaluation of a commercial activity-tracker at slower walking speeds. Gait & posture 59(2018), 177–181.Google Scholar
- Kun Xia, Jianguang Huang, and Hanyu Wang. 2020. LSTM-CNN Architecture for Human Activity Recognition. IEEE Access 8(2020), 56855–56866. https://doi.org/10.1109/ACCESS.2020.2982225Google ScholarCross Ref
- Zhenqi Xu, Shan Li, and Weihong Deng. 2015. Learning temporal features using LSTM-CNN architecture for face anti-spoofing. In 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR). 141–145. https://doi.org/10.1109/ACPR.2015.7486482Google ScholarCross Ref
- Chih-Hsiang Yang and David E Conroy. 2019. Feasibility of an outdoor mindful walking program for reducing negative affect in older adults. Journal of aging and physical activity 27, 1 (2019), 18–27.Google ScholarCross Ref
Index Terms
- Activity Recognition in Older Adults with Training Data from Younger Adults: Preliminary Results on in Vivo Smartwatch Sensor Data
Recommendations
MyMove: Facilitating Older Adults to Collect In-Situ Activity Labels on a Smartwatch with Speech
CHI '22: Proceedings of the 2022 CHI Conference on Human Factors in Computing SystemsCurrent activity tracking technologies are largely trained on younger adults’ data, which can lead to solutions that are not well-suited for older adults. To build activity trackers for older adults, it is crucial to collect training data with them. To ...
Older Adults' Digital Gameplay
Background. Empirical evidence suggests that digital gameplay can enhance social interaction and improve cognition for older adults. However, if digital games are to be effectively used as interventions to address age-related challenges, it is important ...
Recruiting Older Adults in the Wild: Reflections on Challenges and Lessons Learned from Research Experience
PervasiveHealth '18: Proceedings of the 12th EAI International Conference on Pervasive Computing Technologies for HealthcareIt is important to understand the older adults prior to the design process. The understanding can better facilitate design conversations between the researchers and the older adults. In this paper, we discussed our experiences of building a relationship ...
Comments