Author(s):
- Kathy Li
- Iñigo Urteaga
- Amanda Shea
- Virginia J Vitzthum
- Chris H Wiggins
- Noémie Elhadad
Abstract:
Objective
The study sought to build predictive models of next menstrual cycle start date based on mobile health self-tracked cycle data. Because app users may skip tracking, disentangling physiological patterns of menstruation from tracking behaviors is necessary for the development of predictive models.
Materials and Methods
We use data from a popular menstrual tracker (186 000 menstruators with over 2 million tracked cycles) to learn a predictive model, which (1) accounts explicitly for self-tracking adherence; (2) updates predictions as a given cycle evolves, allowing for interpretable insight into how these predictions change over time; and (3) enables modeling of an individual’s cycle length history while incorporating population-level information.
Results
Compared with 5 baselines (mean, median, convolutional neural network, recurrent neural network, and long short-term memory network), the model yields better predictions and consistently outperforms them as the cycle evolves. The model also provides predictions of skipped tracking probabilities.
Discussion
Mobile health apps such as menstrual trackers provide a rich source of self-tracked observations, but these data have questionable reliability, as they hinge on user adherence to the app. By taking a machine learning approach to modeling self-tracked cycle lengths, we can separate true cycle behavior from user adherence, allowing for more informed predictions and insights into the underlying observed data structure.
Conclusions
Disentangling physiological patterns of menstruation from adherence allows for accurate and informative predictions of menstrual cycle start date and is necessary for mobile tracking apps. The proposed predictive model can support app users in being more aware of their self-tracking behavior and in better understanding their cycle dynamics.
Documentation:
https://doi.org/10.1093/jamia/ocab182
References:
1. LiIA stage-based model of personal informatics systemsProceedings of the SIGCHI Conference on Human Factors in Computing Systems2010557566Google Scholar
2. Fox S, Duggan M. Tracking for health. 2013. http://www.pewinternet.org/2013/01/28/tracking-for-health/. Accessed September 2, 2021.
3. McKillop M, Mamykina L, Elhadad N. Designing in the dark: eliciting self-tracking dimensions for understanding enigmatic disease. In: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems; 2018: 565.Google Scholar
4. Costa Figueiredo M, Caldeira C, Reynolds TL, Victory S, Zheng K, Chen Y. Self-tracking for fertility care: collaborative support for a highly personalized problem. Proc ACM Hum-Comput Interact 2017; 1 (CSCW): 1.
5. Ayobi A, Marshall P, Cox AL, Chen Y. Quantifying the body and caring for the mind: self-tracking in multiple sclerosis. In: Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems; 2017: 6889–901.Google Scholar
6. Desai PM, Mitchell EG, Hwang ML, Levine ME, Albers DJ, Mamykina L. Personal health oracle: explorations of personalized predictions in diabetes self-management. In: Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems; 2019: 370. doi:10.1145/3290605.3300600.Google Scholar
7. Shaw RJ, Steinberg DM, Bonnet J, et al. Mobile health devices: will patients actually use them? J Am Med Inform Assoc 2016; 23 (3): 462–6.
8. Vaghefi I, Tulu B. The continued use of mobile health apps: insights from a longitudinal study. JMIR MHealth UHealth 2019; 7 (8): e12983.
9. Choe EK, Lee NB, Lee B, Pratt W, Kientz JA. Understanding quantified-selfers’ practices in collecting and exploring personal data. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems; 2014: 1143–52. doi: 10.1145/2556288.2557372.Google Scholar
10. Consolvo S, Klasnja P, McDonald DW, Landay JA. Designing for healthy lifestyles: design considerations for mobile technologies to encourage consumer health and wellness. Front Hum Comput Interact 2012; 6 (3–4): 167–315.
11. Epstein DA, Ping A, Fogarty J, Munson SA. A lived informatics model of personal informatics. In: Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing; 2015: 731–42. doi: 10.1145/2750858.2804250.Google Scholar
12. Wartella EA, Rideout V, Montague H, Beaudoin-Ryan L, Lauricella AR. Teens, health and technology: a national survey. Media Commun 2016; 4 (3): 13–23.
14. Fox S, Epstein DA. Monitoring menses: design-based investigations of menstrual tracking applications. In: Bobel C, Winkler IT, Fahs B, Hasson KA, Kissling EA, Roberts T-A, eds. Palgrave Handbook of Critical Menstruation Studies. New York, NY: Springer; 2020: 733–50.
15. Epstein DA, Lee NB, Kang JH, et al. Examining menstrual tracking to inform the design of personal informatics tools. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems; 2017; 2017: 6876–88.
16. Pierson E, Althoff T, Leskovec J. Modeling individual cyclic variation in human behavior. In: Proceedings of the 2018 World Wide Web Conference; 2018: 107–16. doi: 10.1145/3178876.3186052.Google Scholar 17. Symul L, Wac K, Hillard P, Salathé M. Assessment of menstrual health status and evolution through mobile apps for fertility awareness. NPJ Digit Med 2019; 2: 64.
18. Bull JR, Rowland SP, Scherwitzl EB, Scherwitzl R, Danielsson KG, Harper J. Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles. NPJ Digit Med 2019; 2: 83.
19.Li K, et al. Characterizing physiological and symptomatic variation in menstrual cycles using self-tracked mobile health data. NPJ Digit Med 2020; 3: 79. doi: 10.1038/s41746-020-0269-8.
20. Pierson E, Althoff T, Thomas D, Hillard P, Leskovec J. Daily, weekly, seasonal and menstrual cycles in women’s mood, behaviour and vital signs. Nat Hum Behav 2021; 5: 716–25.
21. Soumpasis I, Grace B, Johnson S. Real-life insights on menstrual cycles and ovulation using big data. Hum Reprod Open 2020; 2020 (2): hoaa011.doi:10.1093/hropen/
22. Vitzthum VJ. The ecology and evolutionary endocrinology of reproduction in the human female. Am J Phys Anthropol 2009; 140 (Suppl 49): 95–136.
23. Arey LB. The degree of normal menstrual irregularity. Am J Obstet Gynecol 1939; 37 (1): 12–29.
24. Treloar AE, Boynton RE, Behn BG, Brown BW. Variation of the human menstrual cycle through reproductive life. Int J Fertil 1967; 12 (1 Pt 2): 77–126.
25. Chiazze L, Brayer FTJr, Macisco JJ, Parker MP, Duffy BJ. The length and variability of the human menstrual cycle. JAMA 1968; 203 (6): 377–80.
26. Belsey EM, Pinol AP. Menstrual bleeding patterns in untreated women. Task Force on Long-Acting Systemic Agents for Fertility Regulation. Contraception 1997; 55 (2): 57–65.
27. Burkhart MC, de Mazariegos L, Salazar S, Hess T. Incidence of irregular cycles among Mayan women who reported having regular cycles: implications for fertility awareness methods. Contraception 1999; 59 (4): 271–5.
28. Vitzthum VJ, Spielvogel H, Caceres E, Gaines J. Menstrual patterns and fecundity among non-lactating and lactating cycling women in rural highland Bolivia: implications for contraceptive choice. Contraception 2000; 62 (4): 181–7.
29. Creinin MD, Keverline S, Meyn LA. How regular is regular? An analysis of menstrual cycle regularity. Contraception 2004; 70 (4): 289–92.
30. S. R W. Menstrual cycle characteristics and predictability of ovulation of Bhutia women in Sikkim, India. J Physiol Anthropol 2006; 25 (1): 85–90.
34. Ferrell RJ, O’Connor KA, Rodríguez G, et al. Monitoring reproductive aging in a 5-year prospective study: aggregate and individual changes in steroid hormones and menstrual cycle lengths with age. Menopause 2005; 12 (5): 567–757.
35. Harlow SD, Gass M, Hall JE, et al. ; STRAW + 10 Collaborative Group. Executive summary of the stages of reproductive aging workshop + 10: addressing the unfinished agenda of staging reproductive aging. J Clin Endocrinol Metab 2012; 97 (4): 1159–68.
36. Bishop CM. Model-based machine learning. Philos Trans A Math Phys Eng Sci 2013; 371 (1984): 20120222.
37. Koller D, Friedman N. Probabilistic Graphical Models: Principles and Techniques. Cambridge, MA: MIT Press, 2009.Google Scholar
38. Bortot P, Masarotto G, Scarpa B. Sequential predictions of menstrual cycle lengths. Biostatistics 2010; 11 (4): 741–55.
39. Harlow SD, Zeger SL. An application of longitudinal methods to the analysis of menstrual diary data. J Clin Epidemiol 1991; 44 (10): 1015–25.
40. Harlow SD, Lin X, Ho MJ. Analysis of menstrual diary data across the reproductive life span Applicability of the bipartite model approach and the importance of within-woman variance. J Clin Epidemiol 2000; 53 (7): 722–33.
41. Guo Y, Manatunga AK, Chen S, Marcus M. Modeling menstrual cycle length using a mixture distribution. Biostatistics 2006; 7 (1): 100–14.
42. Oliveira T, Bruinvels G, Pedlar C, Newell J. Modelling menstrual cycle length in athletes using state-space models. Sci Rep 2021; 11: 16972. doi: 10.21203/rs.3.rs-122553/v1.
43. Urteaga I, Li K, Shea A, Vitzthum V, Wiggins CH, Elhadad N. A generative modeling approach to calibrated predictions: a use case on menstrual cycle length prediction. In: Proceedings of the 6th Machine Learning for Healthcare Conference; 2021.
44. Shea AA, Vitzthum VJ. The extent and causes of natural variation in menstrual cycles: integrating empirically-based models of ovarian cycling into research on women’s health. Drug Discov Today Dis Models 2020; 32: 41–9.