Causal discovery
Learning plausible causal structure between behaviors, physiology and outcomes.
Go beyond correlation.
Observational wearable data can answer causal questions when the analysis is designed for it. We estimate effects, test mediators and build causal graphs from longitudinal data.
What we deliver
Causal analysis, end to end.
Causal effect estimation
Effects of exposures such as physical activity on outcomes such as quality of life, with causal machine learning.
Mediator analysis
Which pathways carry an effect, for example HRV and activity as mediators of the pandemic’s effect on sleep.
Counterfactual reasoning
What would likely have happened under a different exposure or intervention.
Structural equation modeling
Latent constructs and their relationships, tested against your data.
Factor analysis
Data-driven constructs from questionnaires, EMAs and sensor features.
How we analyze it
- Causal machine learning
- Mediation analysis
- Structural equation and factor models
- Personalized causal graphs
What you receive
- Effect estimates with uncertainty
- Causal diagrams and mediation figures
- Sensitivity analyses and a methods section
- Documented, reproducible code
Evidence
Selected peer-reviewed work by our team.
- Khatibi et al. Impact of COVID-19 pandemic on sleep including HRV and physical activity as mediators: a causal ML approach. IEEE BSN, 2023. doi:10.1109/BSN58485.2023.10331423
- Kazemi et al. Impact of physical activity on quality of life during pregnancy: a causal ML approach. IEEE EMBC, 2024. doi:10.1109/EMBC53108.2024.10782584
- Yang and Rahmani. Personalized causal graph reasoning for LLMs: an implementation for dietary recommendations. IEEE Journal of Biomedical and Health Informatics, 2025. doi:10.1109/JBHI.2025.3624144