Temporal patterns and seasonality
Daily, weekly and longer cycles in sleep, activity, heart rate and EMAs.
Make sense of months of continuous data.
Long studies produce dense, messy time series. We explore them for patterns, changes and outliers, handle missing data, and turn them into findings you can explain.
What we deliver
Longitudinal time-series analytics, end to end.
Change point detection
When a participant’s signals shift, for example across pregnancy, treatment or a life event.
Anomaly detection
Unusual days or readings flagged for review or for intervention.
Clustering and phenotypes
Groups of participants or days with similar patterns.
Event mining
Sequences of events that tend to occur together or precede outcomes.
Missing data and personalization
Methods that stay reliable when wear time and responses are incomplete.
How we analyze it
- Multivariate time-series analysis
- Graph attention networks for personalized change analysis
- Clustering, anomaly and change point algorithms
- Missing data resilient, personalized modeling
What you receive
- Exploratory reports with clear figures
- Participant-level timelines and summaries
- Inputs for causal and predictive models
- Documented, reproducible code
Evidence
Selected peer-reviewed work by our team.
- Wang et al. Personalized graph attention network for multivariate time-series change analysis: a case study on long-term maternal monitoring. ACM SAC, 2023. doi:10.1145/3555776.3577675
- Azimi et al. Missing data resilient decision-making for healthcare IoT through personalization: a case study on maternal health. Future Generation Computer Systems, 2019. doi:10.1016/j.future.2019.02.015
- Sarhaddi et al. Long-term IoT-based maternal monitoring: system design and evaluation. Sensors, 2021. doi:10.3390/s21072281
- Sarhaddi et al. Trends in heart rate and heart rate variability during pregnancy and the 3-month postpartum period: continuous monitoring in a free-living context. JMIR mHealth and uHealth, 2022. doi:10.2196/33458