ABSTRACT
Mobile health (mHealth) technologies enable frequent sampling of physiological and psychological signals over time. In our recent work we used a convolutional neural network (CNN) model to predict self-reported phenotypes of chronic conditions from step and sleep data recorded from passive trackers in free living conditions. We investigated the impact of the time-granularity of the collected data and showed that training the models on higher- resolution (minute-level) data improved classification performance on conditions related to mental health and nervous system disorders, as compared to using only day-level totals. In the present work we shift the focus from the time resolution of the observation window to its duration. We study how the performance of the best-performing model on the highest-resolution data changes as the length of the data collection window is varied from 3 to 147 days for each user. We found that for mental health and nervous system disorders, a model trained on 30 days of mHealth data attains the same performance as using the full 147-day window of data, in terms of AUC increase over a baseline model that uses only demographics, height, and weight. Additionally, for the same cluster of conditions, only 7 days of data are sufficient to realize 62% of the maximum increase in AUC over baseline attainable using the full window. The results suggest that for some conditions health-related digital phenotyping in free-living conditions can potentially be performed in a relatively short amount of time, imposing minimal disruptions on user habits.
- R. Caruana, S. Baluja, T. Mitchell, et al. Using the future to "sort out" the present: Rankprop and multitask learning for medical risk evaluation. In Advances in Neural Information Processing Systems (NIPS) 8, pages 959--965, 1996. Google ScholarDigital Library
- F. Chollet. Keras. https://github.com/fchollet/keras, 2015.Google Scholar
- L. Hood and S. H. Friend. Predictive, personalized, preventive, participatory (p4) cancer medicine. Nature Reviews Clinical Oncology, 8(3):184--187, 2011.Google ScholarCross Ref
- Y. A. LeCun, L. Bottou, G. B. Orr, and K.-R. Müller. Efficient backprop. In Neural networks: Tricks of the trade, pages 9--48. Springer, 2012. Google Scholar
- X. Li, J. Dunn, D. Salins, G. Zhou, W. Zhou, S. M. S.-F. Rose, D. Perelman, E. Colbert, R. Runge, S. Rego, et al. Digital health: Tracking physiomes and activity using wearable biosensors reveals useful health-related information. PLoS Biology, 15(1):e2001402, 2017.Google ScholarCross Ref
- Z. C. Lipton, D. C. Kale, C. Elkan, and R. Wetzel. Learning to diagnose with LS™ recurrent neural networks. In Proceedings of the 2016 International Conference on Learning Representations (ICLR), 2016.Google Scholar
- Precision Medicine Initiative Working Group et al. Report to the advisory committee to the director: The precision medicine initiative cohort program-building a research foundation for 21st century medicine. Washington, DC: National Institutes of Health, 2015.Google Scholar
- Quantified Self Labs. The quantified self. www.quantifiedself.com, 2016. Accessed: 2016-05-20.Google Scholar
- T. Quisel, D. C. Kale, and L. Foschini. Intra-day activity better predicts chronic conditions. In 30th Conference on Neural Information Processing Systems (NIPS 2016), 2016.Google Scholar
- B. Ramsundar, S. Kearnes, P. Riley, D. Webster, D. Konerding, and V. Pande. Massively multitask networks for drug discovery. arXiv preprint arXiv:1502.02072, 2015.Google Scholar
- P. Schulam and S. Saria. A framework for individualizing predictions of disease trajectories by exploiting multi-resolution structure. In Advances in Neural Information Processing Systems, pages 748--756, 2015. Google ScholarDigital Library
- Y. Zheng, Q. Liu, E. Chen, Y. Ge, and J. L. Zhao. Time series classification using multi-channels deep convolutional neural networks. In Web-Age Information Management, pages 298--310. Springer, 2014.Google Scholar
- A. Zikeba and P. Ramza. Standard deviation of the mean of autocorrelated observations estimated with the use of the autocorrelation function estimated from the data. Metrology and Measurement Systems, 18(4):529--542, 2011.Google Scholar
Index Terms
- Observation Time vs. Performance in Digital Phenotyping
Recommendations
A survey on big data-driven digital phenotyping of mental health
Highlights- The research about the digital phenotyping of mental health is limited.
- ...
AbstractThe landscape of mental health has undergone tremendous changes within the last two decades, but the research on mental health is still at the initial stage with substantial knowledge gaps and the lack of precise diagnosis. Nowadays, ...
A comparative analysis of sepsis digital phenotyping methods
ACSW '21: Proceedings of the 2021 Australasian Computer Science Week MulticonferenceHealth data captured in Electronic health records (EHRs) have enabled the development of computational approaches to improve patient management and treatment, including early diagnosis of severe conditions such as sepsis. The validity of these efforts, ...
I will prescribe you an app
SummerSim '14: Proceedings of the 2014 Summer Simulation MulticonferenceMedical applications are used as medical information on 24 %, 22% are dedicated to the monitoring of physical parameters, 18 % easier to track disease, 16% for education and management, and 6% respectively aid to diagnosis.
These tools represent an ...
Comments