THESSALONIKI, Greece – Researchers at the Artistotle University of Thessaloniki are developing a set of technology-based solutions for the early detection and care of Parkinson’s disease through the iPrognosis project. The project is based on the unobtrusive collection of behavioral data from users’ natural interaction with their smart devices that may be linked to early Parkinson’s symptoms. The team launched the iPrognosis mobile app, which collects a variety of data including voice characteristics while users are talking on the phone; hand steadiness while they’re holding the device; and keystrokes-related data when using the app’s keyboard, in 2017 in Germany, Greece, Portugal and the U.K. Other information is also gathered about distance covered each day, facial expressions from stored photos and emotional content from stored text messages. Data from smartwatch heart rate and skin temperature sensors are also being used to monitor sleep quality, since sleep disorders are an early symptom of Parkinson’s. So far, around 433,625 records—about 90 GB of data—have been collected to develop machine-learning algorithms that can detect Parkinson’s-related behavioral changes. The iPrognosis project is currently capturing additional data on food consumption rates, bowel sounds and heart rates from plate scales, smart belts and smart television remote controls. Ultimately, iPrognosis plans to design interventions to help Parkinson’s patients sustain their quality of life, in collaboration with their doctors.
You are here: / / Project collects smart devices to link behavior to early Parkinson’s symptoms