gms | German Medical Science

64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

08. - 11.09.2019, Dortmund

Smartwatch-based Examination of Movement Disorders: Early Implementation and Measurement Accuracy

Meeting Abstract

  • Julian Varghese - University of Münster, Münster, Germany
  • Stephan Niewöhner - University of Münster, Münster, Germany
  • Michael Fujarski - University of Münster, Münster, Germany
  • Iñaki Soto-Rey - University of Münster, Münster, Germany
  • Anna-Lena Schwake - University Hospital of Münster, Münster, Germany
  • Tobias Warnecke - University Hospital of Münster, Münster, Germany
  • Martin Dugas - University of Münster, Münster, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Dortmund, 08.-11.09.2019. Düsseldorf: German Medical Science GMS Publishing House; 2019. DocAbstr. 52

doi: 10.3205/19gmds136, urn:nbn:de:0183-19gmds1368

Published: September 6, 2019

© 2019 Varghese et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



Introduction: Movement Disorders such as Parkinson’s disease are primarily diagnosed via neurological examination and there is a strong need for objective measures and early diagnosis [1], [2]. Establishing a technology-driven two-year prospective study, we implemented a novel mobile system with integration of smartwatches into neurological examination of patients with movement disorders and healthy controls [3]. The aim of the system is to establish an objective measuring framework to analyze hand tremor and other potential phenotypical biomarkers, which can provide predictive support in early diagnosis. The following work presents details on the running system, the established research platform and measurement accuracy.

Methods: The iOS-based system utilizes two synchronous Apple watches. Each examination consists of 10 assessment steps, which were designed with Movement Disorders specialists. They aim to systematically provoke and monitor a set of movement characteristics as tremor occurrence and bradykinesia. In addition, a paired smartphone guides the examiner through the examination steps and enables questionnaire-based data capture of early symptoms, medication and family history. All captured data is sent to a privacy-preserving research database with a self-implemented analytics platform. It enables visualization of acceleration, rotation, questionnaires data and comparison of different participant population regarding diagnosis and age.

Acceleration data is going to be analyzed with Fast Fourier Transform to infer dominant tremor frequency in each examination section. Distance of movement amplitude is calculated via double integration of the signal that was corrected by using a band-pass filter and inverse FFT. Accuracy of tremor amplitude and dominant frequencies were evaluated by systematic comparisons (n=100) with an Android-based reference application, called LiftPulse for four different clinically relevant amplitude and frequency ranges [4].

Results: As of today, 182 participants were enrolled and measured. Among of those, 118 have Parkinson’s disease based on confirmation by movement disorder experts. Twenty-four have other movement disorders, 40 participants are healthy controls. The data visualization platform is readily implemented and data from all enrolled study participants could be imported and visualized successfully.

Regarding measurement accuracy, distance deviation of tremor amplitude was in all measurements <0,5 mm and all (100%) dominant frequencies of each test-measurement were mapped to the correct clinically relevant frequency intervals. Details with boxplots analysis are provided in [4].

Discussion: We have established a novel mobile system that prospectively collects accurate measurement data at both body sides. A visualization platform enables manual comparisons of different participant subpopulations. In some cases of the running study, even subtle tremor occurrences with Amplitude <0.02 G could be captured in early Parkinson’s diagnosis, which are not detectable by human vision. These observations indicate the potential for spotting subtle movement characteristics, which can contribute to yet-unidentified movement patterns. We expect at least 500 participants by the end of the study (October 2020). The highly structured, two-hands-based data capture is unique for all movement disorders and is building the foundation to train a Machine Learning classifier for subsequent disease prediction.

Registration on German Clinical Trials Register: DRKS00016594.

The authors declare that they have no competing interests.

The authors declare that a positive ethics committee vote has been obtained.


Rizzo G, Copetti M, Arcuti S, Martino D, Fontana A, Logroscino G. Accuracy of clinical diagnosis of Parkinson disease: A systematic review and meta-analysis. Neurology. 2016; 86(6):566–76.
Espay AJ, Bonato P, Nahab FB, Maetzler W, Dean JM, Klucken J, Eskofier BM, et al. Technology in Parkinson's disease: Challenges and opportunities. Mov Disord. 2016; 31(9):1272–82.
Varghese J, Niewöhner S, Soto-Rey I, Schipmann-Miletić S, Warneke N, Warnecke T, Dugas M. A Smart Device System to Identify New Phenotypical Characteristics in Movement Disorders. Front. Neurol. 2019; 10:5.
Julian Varghese. Online Supplement Smart Device System. [Accessed 2019 Jan 3]. Available from: URL: External link