ActiveAgeing – The DIGI.PARK branch

PD is characterized by disturbances in motor behavior, including tremors, slowness, stiffness, and several other problems due to a degeneration of neural pathways for which there are no biomarkers, making diagnosis and research challenging. The symptoms are challenging to measure over time due to subjective and low-resolution assessment methods. Current tools for assessing clinical phenotypes and severity of Parkinson’s disease (PD) are based on observation while the patient performs a series of tasks. The Unified Parkinson's Disease Rating Scale (UPDRS) is considered the gold standard for assessing the efficacy of clinical trials testing symptomatic and neuroprotective agents. These tools are however limited by lack of objectiveness, low sensitivity and reproducibility, and vast variations depending on the time of the examination, time of last received dose of dopaminergic treatment, etc. One approach to circumvent these limitations and establish more objective measures of severity is that of digital phenotyping via the use of wearable sensor devices. 

The aim of the DIGI.PARK study is to explore the use of wearable sensors for symptom tracking in home-dwelling people with Parkinson’s disease. 

This branch of ActiveAgeing is an observational study comprised of two phases. In the first phase, we investigated the use of wearable sensor technology for research and clinical work in Parkinson’s disease. A 2-week data collection was conducted in the participants’ homes on 15 participants with Parkinson’s and 15 participants without Parkinson’s from the innovative living environment Helgetun. We employed clinical assessment tools (cognitive assessment, parkinsonian symptomology, sleep disturbances), two smartwatches (Fitbit Sense and Empatica E4), and a smart ring (Oura). 

The second phase of the study is based on the results of the first phase, as the data collection procedure is refined according to the first-phase data analysis. The second phase involves data collection from persons with Parkinson’s disease and their spouses, to compare the crossover effects of the disease. Both phases include the design of specific Parkison’s disease digital biomarkers for symptom tracking. 

Part of a two-sided study

The ActiveAgeing study consists of two branches – the DIGI.PARK and the Helgetun branch. The Helgetun branch is exploring how living in an innovative, community-based environment can affect the lives of older adults, using a qualitative approach. See separate description on the Helgetun branch here.

Team

DIGI.PARK is a collaborative initiative between the Centre for Clinical Treatment Research Neuro-SysMed and SEFAS. Our team comprises Haakon Reithe, PhD student and main investigator of the project, Dr. Monica Patrascu, systems engineer, Dr. Juan Carlos Torrado Vidal, computer engineer, Dr. Brice Marty, electrical engineer and neuroscientist, Elise Førsund, PhD student in the Helgetun branch of ActiveAgeing, Professor Bettina Husebø, anesthesiologist and leader of SEFAS, and Professor Charalampos Tzoulis, neurologist and co-leader of Neuro-SysMed.

Status

The first phase study was initiated in the spring of 2021 and all data was collected in 2021/2022. The comparative cross-correlation analysis of the three wearable devices is finalized, and a digital biomarker for tremor quantification and a digital biomarker for physical activity response are designed. 

PhD student Haakon Reithe submitted the first paper in the fall of 2024, where we cross-evaluated devices for Parkinson’s research and clinical use. Our findings indicate that commercial-grade and research-grade devices alone are not optimal. Research-grade devices provide excellent data resolution and access, whereas commercial-grade devices are user-friendly both for researchers and participants. We conclude that for symptom tracking, gaining access to raw data from user-friendly sensors is essential for scaling efforts in decentralized PD research. 

Reithe is currently finishing the second paper, with work testing an algorithm that quantifies the intensity of tremors in ranges of 3 to 12 Hz. The algorithm was developed together with Monica Patrascu and Brice Marty, and in the paper, we validate the algorithm on participants who exhibit unilateral tremors by examining the differences between the most tremor-affected hand with the least affected hand. The algorithm output gives us a tremor index (TI). Preliminary results indicate that the TI of the most tremor-affected hand is higher than the least affected hand. 

For the third paper, we are making of the TI to evaluate the response to medication, by comparing computed TI pre- and post-medication. In this paper, we will also include control participants, allowing us to test both between and within participants. 

Impact

This project investigates how technology can assist society in addressing the challenges posed by an increasingly aging population with Parkinson’s disease (PD). This provides important knowledge for planning of future health care, which is transferrable to other disease that share symptomology with PD, such as essential tremor and other motor dysfunction.

See previous article from the startup of the project.