PERFECT: Personalized Exercise Recommendation Framework and architECTure
Authors: 
Milad Asgari Mehrabadi, Elahe Khatibi, Tamara Jimah, Sina Labbaf, Holly Borg, Laura Narvaez, Pamela Pimentel, Arlene Turner, 
Nikil Dutt, Yuqing Guo, Amir M. RahmaniAuthors Info & Claims
Milad Asgari Mehrabadi, Elahe Khatibi, Tamara Jimah, Sina Labbaf, Holly Borg, Laura Narvaez, Pamela Pimentel, Arlene Turner, Nikil Dutt, Yuqing Guo, Amir M. RahmaniAuthors Info & ClaimsArticle No.: 25, Pages 1 - 25
Abstract
Background: The health benefits of regular physical activity (PA) are well-established and widely acknowledged. Through the integration of wearable trackers, the Internet of Things (IoT)—a network of interconnected devices capable of collecting and exchanging data—coupled with mobile health (mHealth), which refers to the use of mobile devices to support medical and public health practices, it is now feasible to systematically gather and present individual exercise behaviors. This advanced approach enables the precise correlation of users’ physiological data and daily activities with their specific fitness needs, offering a personalized pathway to improving health outcomes.
Objective: This study aims to enhance PA levels among individuals by developing a personalized exercise recommendation system. Utilizing reinforcement learning, the system proposes tailored exercise plans based on biomarkers and the user’s specific context.
Methods: In this study, we developed applications for smartphones and smartwatches designed to gather, monitor, and recommend exercise routines through the application of a contextual multi-arm bandit algorithm. To evaluate the efficacy of this mHealth exercise regimen, we enlisted the participation of twenty female college students.
Results: The outcomes of our investigation revealed a significant enhancement in the average daily duration of exercise (P \({\lt}\). 001). Participants expressed high levels of satisfaction with both the walking program and the recommendation system, achieving average ratings of 4.31 (SD \(=\) 0.60) and 3.69 (SD \(=\) 0.95), respectively, on a 5-point scale. Furthermore, the average scores for participants’ confidence in safely performing the recommended walking exercises, as well as their perception of the study’s effectiveness in meeting their PA needs, were both above 4, indicating a positive reception and confidence in the program’s design and implementation.
Conclusions: The evolution of the IoT and wearable technology has marked the beginning of a new era for mHealth systems, particularly in the personalization of health interventions. Such advancements enable the precise personalization of PA recommendations, potentially enhancing user engagement and performance outcomes. This paper introduces a novel exercise recommendation system that utilizes reinforcement learning to personalize walking exercises based on the user’s biomarkers and context, aiming to improve the user’s aerobic capacity significantly.
References
[1]
Scott J. Strath, Leonard A. Kaminsky, Barbara E. Ainsworth, Ulf Ekelund, Patty S. Freedson, Rebecca A. Gary, Caroline R. Richardson, Derek T. Smith, Ann M. Swartz, and American Heart Association Physical Activity Committee of the Council on Lifestyle and Cardiometabolic Health and Cardiovascular, Exercise, Cardiac Rehabilitation and Prevention Committee of the Council on Clinical Cardiology, and Council. 2013. Guide to the assessment of physical activity: Clinical and research applications: A scientific statement from the American Heart Association. Circulation 128, 20 (2013), 2259–2279.
[2]
American College of Sports Medicine. 2018. ACSM’s Guidelines for Exercise Testing and Prescription (10th ed.). Lippincott Williams & Wilkins, Philadelphia, PA.
[3]
Jennifer L. Scheid, and Emma O’Donnell. 2019. Revisiting heart rate target zones through the lens of wearable technology. ACSM’s Health & Fitness Journal 23, 3 (2019), 21–26.
[4]
Yangyi Lu, Ziping Xu, and Ambuj Tewari. 2021. Bandit algorithms for precision medicine. arXiv:2108.04782. Retrieved from https://arxiv.org/abs/2108.04782
[5]
Mawulolo K. Ameko, Miranda L. Beltzer, Lihua Cai, Mehdi Boukhechba, Bethany A. Teachman, and Laura E. Barnes. 2020. Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Conference on Recommender Systems.
[6]
Alberto Bietti, Alekh Agarwal, and John Langford. 2021. A contextual bandit bake-off. The Journal of Machine Learning Research 22, 1 (2021), 5928–5976.
[7]
Tobias Schnabel, Adith Swaminathan, Ashudeep Singh, Navin Chandak, and Thorsten Joachims. 2016. Recommendations as treatments: Debiasing learning and evaluation. In Proceedings of the 33rd International Conference on Machine Learning. PMLR.
[8]
Peng Liao, Kristjan Greenewald, Predrag Klasnja, and Susan Murphy. 2020. Personalized heartsteps: A reinforcement learning algorithm for optimizing physical activity. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 4, 1 (2020), 1–22.
[9]
N. Bolger and J. P. Laurenceau. 2013. Intensive Longitudinal Methods: An Introduction to Diary and Experience Sampling Research. Guilford Press.
[10]
S. W. Raudenbush and A. S. Bryk. 2002. Hierarchical Linear Models: Applications and Data Analysis Methods (2nd ed.). Sage, Thousand Oaks, CA.
[11]
I. Adjerid, G. Loewenstein, R. Purta, and A. Striegel. 2022. Gain-loss incentives and physical activity: The role of choice and wearable health tools. Management Science 68, 4 (2022), 2642–2667.
[12]
M. Milne-Ives, C. Lam, C. De Cock, M.H. Van Velthoven, and E. Meinert. 2020. Mobile apps for health behavior change in physical activity, diet, drug and alcohol use, and mental health: Systematic review. JMIR mHealth and uHealth 8, 3 (2020), e17046.
[13]
J. Fang, V. Lee, and H. Wang. 2022. Dynamic physical activity recommendation on personalised mobile health information service: A deep reinforcement learning approach. arXiv:2204.00961. Retrieved from https://arxiv.org/abs/2204.00961
[14]
M. K. Ameko, M. L. Beltzer, L. Cai, M. Boukhechba, B. A. Teachman, and L. E. Barnes. 2020. Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Conference on Recommender Systems, 249–258.
[15]
E. Yom-Tov, G. Feraru, M. Kozdoba, S. Mannor, M. Tennenholtz, and I. Hochberg. 2017. Encouraging physical activity in patients with diabetes: Intervention using a reinforcement learning system. Journal of Medical Internet Research 19, 10 (2017), e7994.
[16]
M. Rabbi, M. H. Aung, M. Zhang, and T. Choudhury. 2015. MyBehavior: Automatic personalized health feedback from user behaviors and preferences using smartphones. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 707–718.
[17]
M. Saponaro, A. Vemuri, G. Dominick, and K. Decker. 2021. Contextualization and individualization for just-in-time adaptive interventions to reduce sedentary behavior. In Proceedings of the Conference on Health, Inference, and Learning, 246–256.
[18]
S. Wang, K. Sporrel, H. van Hoof, M. Simons, R.D. de Boer, D. Ettema, N. Nibbeling, M. Deutekom, and B. Kröse. 2021. Reinforcement learning to send reminders at right moments in smartphone exercise application: A feasibility study. International Journal of Environmental Research and Public Health 18, 11 (2021), 6059.
[19]
A. Mahyari and P. Pirolli. 2021. Physical exercise recommendation and success prediction using interconnected recurrent neural networks. In Proceedings of the IEEE International Conference on Digital Health (ICDH). IEEE, 148–153.
[20]
M. K. Ameko, M. L. Beltzer, L. Cai, M. Boukhechba, B. A. Teachman, and L. E. Barnes. 2020. Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Conference on Recommender Systems, 249–258.
[21]
M. L. Beltzer, M. K. Ameko, K. E. Daniel, A. R. Daros, M. Boukhechba, L. E. Barnes, and B. A. Teachman. 2022. Building an emotion regulation recommender algorithm for socially anxious individuals using contextual bandits. British Journal of Clinical Psychology 61 (2022), 51–72.
[22]
P. Klasnja, S. Smith, N. J. Seewald, A. Lee, K. Hall, B. Luers, E. B. Hekler, and S. A. Murphy. 2019. Efficacy of contextually tailored suggestions for physical activity: A micro-randomized optimization trial of HeartSteps. Annals of Behavioral Medicine 53, 6 (2019), 573–582.
[23]
Y. Lu, Z. Xu, and A. Tewari. 2021. Bandit algorithms for precision medicine. arXiv:2108.04782. Retrieved from https://arxiv.org/abs/2108.04782
[24]
Eszter Füzéki and Winfried Banzer. 2018. Physical activity recommendations for health and beyond in currently inactive populations. The International Journal of Environmental Research and Public Health 15, 5 (2018), 1042.
[25]
Philipe de Souto Barreto. 2015. Global health agenda on non-communicable diseases: Has WHO set a smart goal for physical activity? BMJ 350 (2015), h23.
[26]
M. Miyachi, J. Tripette, R. Kawakami, and H. Murakami. 2015. “+10 min of physical activity per day”: Japan is lookingfor efficient but feasible recommendations for its population. Journal of Nutritional Science and Vitaminology 61 (2015), S7–S9.
[27]
Mo Zhou, Yoshimi Fukuoka, Yonatan Mintz, Ken Goldberg, Philip Kaminsky, Elena Flowers, and Anil Aswani. 2018. Evaluating machine learning–based automated personalized daily step goals delivered through a mobile phone app: Randomized controlled trial. JMIR mHealth and uHealth 6, 1 (2018), e9117.
[29]
Haskell, L. William, J. Montoye Henry, and Diane Orenstein. 1985. Physical activity and exercise to achieve health-related physical fitness components. Public Health Reports 100, 2 (1985), 202.
[30]
Labbaf Sina, Mahyar Abbasian, Iman Azimi, Nikil Dutt, and Amir M. Rahmani. “ZotCare: a flexible, personalizable, and affordable mhealth service provider.” Frontiers in Digital Health 5 (2023), 1253087.
[31]
Paul A. Harris, Robert Taylor, Brenda L. Minor, Veida Elliott, Michelle Fernandez, Lindsay O’Neal, Laura McLeod, Giovanni Delacqua, Francesco Delacqua, and Jacqueline Kirby. 2019. The REDCap consortium: Building an international community of software platform partners. Journal of Biomedical Informatics 95 (2019), 103208.
[32]
Paul A. Harris, Robert Taylor, Robert Thielke, Jonathon Payne, Nathaniel Gonzalez, and Jose G. Conde. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics 42, 2 (2009), 377–381.
Appendix
PA intensity: low 1 (\({{\lt}}{=}30\) min), moderate 2 (31–60 min), vigorous 3 (\({\gt}\)60 min)
Low PA intensity for 20 subjects
Moderate PA intensity for 20 subjects
Vigorous PA intensity for 20 subjects
12 subjects
Low PA intensity for 12 subjects
Moderate PA intensity for 12 subjects
Vigorous PA intensity for 12 subjects
8 subjects
Low PA intensity for 8 subjects
Moderate PA intensity for 8 subjects
Vigorous PA intensity for 8 subjects
Fisher’s test comparing intensity by subjects included/excluded in analysis
Low: 0.04
Moderate: 0.56
Vigorous: 1.0
Exit survey (n = 12)
Satisfaction with HH walking recommendations.
Ability to meet the physical activity needs: 4.27 (0.47)
Enjoy HH watch: 3.36 (1.43)
Exit survey (n = 8)
Walking: 4.4 (0.89)
Ability to meet the physical activity needs: 4.2 (0.84)
Normal Q-Q plot for normality of the residuals for minutes in light/moderate-intensity exercise.
Normal Q-Q plot for normality of the residuals for the weekly exercise duration.
Index Terms
- PERFECT: Personalized Exercise Recommendation Framework and architECTure
Recommendations
Activity Recommendation with Partners
Recommending social activities, such as watching movies or having dinner, is a common function found in social networks or e-commerce sites. Besides certain websites which manage activity-related locations (e.g., foursquare.com), many items on product ...
Parallel Ratio Based CF for Recommendation System
ICCCNT '16: Proceedings of the 7th International Conference on Computing Communication and Networking TechnologiesWith the increase in E-commerce, Recommendation Systems are getting popular to provide recommendations of various items (movies, books, music) to users. To build the Recommendation System (RS), Collaborative Filtering (CF) techniques are proven ...
WeHeart: A Personalized Recommendation Device for Physical Activity Encouragement and Preventing “Cold Start” in Cardiac Rehabilitation
Human-Computer Interaction – INTERACT 2023AbstractThis paper highlights the importance of physical activity in cardiac rehabilitation as a means of reducing morbidity and mortality rates associated with cardiovascular disease. However, forming physical activity habits is a challenge, and the ...
Comments
Information & Contributors
Information
Published In
October 2024
195 pages
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 12 November 2024
Online AM: 03 October 2024
Accepted: 27 August 2024
Revised: 30 June 2024
Received: 26 September 2023
Published in HEALTH Volume 5, Issue 4
Check for updates
Author Tags
Qualifiers
- Research-article
Funding Sources
- Academy of Finland through the SLIM Project
- U.S. National Science Foundation (NSF) through the UNITE
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 172Total Downloads
- Downloads (Last 12 months)172
- Downloads (Last 6 weeks)86
Reflects downloads up to 25 Dec 2024
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in