Science & evidence

Credibility you can cite.

Peer-reviewed publications, and research conducted with hospitals across Belgium as sites of the KC-QI-2023 study. The evidence behind the measures, in the open.

The measures

From raw sensor data to validated kinematic measures.

Parkiwatch applies validated signal processing and statistical learning algorithms to wrist accelerometer data, producing continuous, longitudinal kinematic measures for healthcare professionals.

01 · CAPTURE

Passive sensing

A smartwatch or wrist-worn sensor records triaxial accelerometer data during daily life.

02 · MEASURE

Validated kinematic measures

Signal becomes three validated kinematic measures — tremor, bradykinesia and dyskinesia — calculated alongside supportive digital health measures of typing, mealtime movement, sleep and activity.

03 · REPORT

Dashboard & PDF reports

Measures are made available to healthcare professionals at 15-minute resolution, to be interpreted within the context of standard clinical assessment.

In research · not part of the CE-marked device

The Q-index

An exploratory composite of digital measures under investigation as an indicator of quality-of-life trend. It is a research measure and is not an output of the PARKIWATCH medical device.

Research impact

The science predates the company.

Koios grew out of a decade of published research on digital biomarkers. The founding team and advisors wrote the methods the platform builds on — measured here in citations, not claims.

9,700+combined citations across the team
16selected publications & proceedings
5research areas, one platform
KK

Konstantinos Kyritsis, PhD

CEO
541Citations
13h-index
16i10-index
Signal processingMachine learningSensorsBehavioural modellingComputer vision
Most cited

Detecting parkinsonian tremor from IMU data collected in-the-wild using deep multiple-instance learning

2019 · 81 citations
DI

Dimitris Iakovakis, PhD

CTO
797Citations
16h-index
20i10-index
Biomedical engineeringHuman–computer interactionMachine learningDigital health
Most cited

Touchscreen typing pattern analysis for detecting fine motor skills decline in early-stage Parkinson’s disease

2018 · 92 citations
VP

Vasileios Papapanagiotou, PhD

Senior R&D engineer · Professor @Karolinska
470Citations
13h-index
18i10-index
Digital signal processingMachine learningWearable sensorsEating behaviour analysis
Most cited

A novel chewing detection system based on PPG, audio, and accelerometry

2016 · 89 citations
GG

Prof. MD Gaëtan Garraux

Clinical lead · Neurology professor @CHU Liège
5,473Citations
43h-index
71i10-index
Movement disordersNeurologyClinical researchParkinson’s disease
Most cited

Brain energy metabolism and dopamine function in Parkinson’s disease

2017 · 245 citations
GS

Gregor Strobbe, PhD

Advisor
639Citations
13h-index
16i10-index
EEGMRIBiomarkersEntrepreneurship
Most cited

Digital health solutions for improved patient care and monitoring

2020 · 125 citations
LK

Lampros Kourtis, PhD

Advisor
1,828Citations
23h-index
37i10-index
Digital biomarkersAlzheimer’s diseaseWearable technology
Most cited

Digital biomarkers for Alzheimer’s disease: the mobile/wearable devices opportunity

2019 · 359 citations

Citation counts, h-index and i10-index from the team’s public Google Scholar profiles — figures evolve as the literature does.

Publications

Peer-reviewed & presented.

From touchscreen typing to in-the-wild tremor detection — the team’s work, by research area.

2022

Parkinson’s Disease Detection Based on Running Speech Data From Phone Calls

Laganas, C., Iakovakis, D., Hadjidimitriou, S., Charisis, V., Bostantzopoulou, S., Katsarou, Z., & Hadjileontiadis, L. J.
IEEE Transactions on Biomedical Engineering Parkinson’s disease
2021

DeepFoG: An IMU-based Detection of Freezing-of-Gait Episodes in Parkinson’s Disease Patients via Deep Learning

Bikias, T., Iakovakis, D., Hadjidimitriou, S., Charisis, V., & Hadjileontiadis, L. J.
Frontiers in Robotics and AI — Sensor Fusion and Machine Perception Parkinson’s disease
2021

Assessment of real-life eating difficulties in Parkinson’s disease patients by measuring plate-to-mouth movement elongation with inertial sensors

Kyritsis, K., Fagerberg, P., Ioakimidis, I., Chaudhuri, K., Reichmann, H., Klingelhoefer, L., & Delopoulos, A.
Scientific Reports, 11(1), 1-14 Parkinson’s disease
2020

Screening of Parkinsonian subtle fine-motor impairment from touchscreen typing via deep learning

Iakovakis, D., Hadjidimitriou, S., Charisis, V., Bostantzopoulou, S., Katsarou, Z., Klingelhoefer, L., Chaudhuri, K., & Hadjileontiadis, L. J.
Scientific Reports, 10(1), 1-13 Parkinson’s disease
2020

Unobtrusive detection of Parkinson’s disease from multi-modal and in-the-wild sensor data using deep learning techniques

Papadopoulos, A., Iakovakis, D., Klingelhoefer, L., Bostantjopoulou, S., Chaudhuri, K. R., Kyritsis, K., … & Delopoulos, A.
Scientific Reports, 10(1), 1-13 Parkinson’s disease
2020

Detection of Mild Cognitive Impairment through natural language and touchscreen typing processing

Ntracha, A., Iakovakis, D., Hadjidimitriou, S., Charisis, V., Tsolaki, M., & Hadjileontiadis, L. J.
Frontiers in Digital Health, 2, 19 Mild cognitive impairment
2019

Detecting parkinsonian tremor from IMU data collected in-the-wild using deep multiple-instance learning

Papadopoulos, A., Kyritsis, K., Klingelhoefer, L., Bostanjopoulou, S., Chaudhuri, K. R., & Delopoulos, A.
IEEE Journal of Biomedical and Health Informatics, 24(9), 2559-2569 Parkinson’s disease 81 citations
2019

Touchscreen typing pattern analysis for remote detection of the depressive tendency

Mastoras, R. E., Iakovakis, D., Hadjidimitriou, S., Charisis, V., Kassie, S., Alsaadi, T., & Hadjileontiadis, L. J.
Scientific Reports, 9(1), 1-12 Depression
2018

Touchscreen typing-pattern analysis for detecting fine motor skills decline in early-stage Parkinson’s disease

Iakovakis, D., Hadjidimitriou, S., Charisis, V., Bostantzopoulou, S., Katsarou, Z., & Hadjileontiadis, L. J.
Scientific Reports, 8(1), 1-13 Parkinson’s disease 92 citations
2018

Motor Impairment Estimates via Touchscreen Typing Dynamics Toward Parkinson’s Disease Detection From Data Harvested In-the-Wild

Iakovakis, D., Hadjidimitriou, S., Charisis, V., Bostantzopoulou, S., Katsarou, Z., Klingelhoefer, L., … & Hadjileontiadis, L. J.
Frontiers in ICT, 5, 28 Parkinson’s disease
2018

Control of eating behavior using a novel feedback system

Esfandiari, M., Papapanagiotou, V., Diou, C., Zandian, M., Nolstam, J., & Södersten, P.
Journal of Visualized Experiments: JoVE, 57432 Eating behaviour 25 citations
2018

Automatic analysis of food intake and meal microstructure based on continuous weight measurements

Papapanagiotou, V., Diou, C., Ioakimidis, I., Södersten, P., & Delopoulos, A.
IEEE Journal of Biomedical and Health Informatics, 23(2), 893-902 Eating behaviour 24 citations
2017

Chewing detection from an in-ear microphone using convolutional neural networks

Papapanagiotou, V., Diou, C., & Delopoulos, A.
39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) Eating behaviour 37 citations
2016

A novel chewing detection system based on PPG, audio, and accelerometry

Papapanagiotou, V., Diou, C., Zhou, L., van den Boer, J., Mars, M., & Ioakimidis, I.
IEEE Journal of Biomedical and Health Informatics, 21(3), 607-618 Eating behaviour 89 citations
2016

Fuzzy logic-based risk of fall estimation using smartwatch data as a means to form an assistive feedback mechanism in everyday living activities

Iakovakis, D., Papadopoulou, F. A., & Hadjileontiadis, L. J.
Healthcare Technology Letters, 3(4), 263-268 Orthostatic hypotension
2016

Standing hypotension prediction based on smartwatch heart rate variability data: a novel approach

Iakovakis, D., & Hadjileontiadis, L. J.
Proceedings of the 18th International Conference on Human-Computer Interaction Orthostatic hypotension
Partners

Research sites and backers.

Research sites · KC-QI-2023
AZ DeltaAZ GroeningeAZ OostendeCHU LiègeCHU Charleroi

Their listing does not indicate routine clinical use of Parkiwatch at these sites.

Backed by
imec.istartVLAIO

Want the methods in detail?

We’re glad to share validation data and discuss study collaborations.

Contact the team → For research partners