Characterizing Behavioral Health Trends through GPS Trajectory Analytics

The Solution

Big Blue AI developed a pioneering behavioral intelligence platform to empower healthcare professionals with objective data, aiming to monitor and improve the daily quality of life for individuals in the early stages of Parkinson’s disease.

• Geospatial & GIS Enrichment: Engineered automated pipelines utilizing reverse geocoding tools and GIS source APIs to determine precise place types, neighborhood classifications, and socioeconomic environments.
• Unsupervised Longitudinal Modeling: Built a sophisticated clustering framework that processes longitudinal location histories to detect subtle behavioral changes over time without requiring pre-labeled training data.
• Mobility Mode Classification: Implemented spatial metrics to automatically segment daily movement into specific transport modes (e.g., walking, driving, public transit) and calculate precise home vs. social space balance.
• Socioeconomic Behavior Embeddings: Deployed cutting-edge language modeling and NLP techniques on spatial trajectories to extract proxy signals regarding patient socioeconomic habits and community engagement.

The Challenge

Managing neurodegenerative conditions like early-stage Parkinson's disease is highly challenging, as changes in a patient's daily quality of life and physical "lifespace" happen gradually and are difficult to track. Koios Care faced the complex hurdle of transforming highly unstructured, noisy, raw GPS time-series data into objective, clinically meaningful biomarkers. Traditional health analytics lacked the tools to map daily movement patterns, such as community interactions, isolation markers, or changes in transportation habits, into an actionable, unified framework that medical professionals could use to evaluate patient well-being and disease progression.

The Impact

The deployment of this pioneering spatial data platform delivered immediate, profound capabilities to the digital health space, earning a Gold Award at the AI & Data Awards 2026:

• 100% Automated Behavioral Benchmarking: Established an adaptable, self-learning "Social Activity Index" that maps complex human mobility patterns without relying on rigid, predefined categories.
• Early Deterioration Detection: Unlocked the power to flag critical behavioral shifts early, such as reduced overall mobility or sudden social isolation, signaling potential health drops before they become emergencies.
• Data-Driven Intervention Customization: Empowered care teams to design highly tailored patient support programs aimed directly at boosting active community participation and targeted physical mobility.
• Catalyzed Public Health Research: Provided a reliable, privacy-preserving data asset to fuel academic research and guide future public policies centered around active, healthy aging.

How Big Blue AI helped 

Big Blue AI acted as a vital strategic partner throughout this collaboration, bridging the gap between raw data science and real-world clinical utility. Our team took complete ownership of the end-to-end algorithmic architecture, guiding the project from initial research and GIS data enrichment to the development of sophisticated, unsupervised machine learning models. By successfully converting noisy, unstructured spatial telemetry into structured behavioral signals, we transformed complex data into a powerful tool for health monitoring. This close, hands-on collaboration not only solved the immediate technical challenges, but also equipped Koios Care with a scalable, future-proof framework ready to redefine digital therapeutics and patient well-being.