Dominik Faißt

Advisors:
Nils Roth (M.Sc.), Dr. med. Franz Marxreiter, Prof. Dr. Björn Eskofier

Duration: 
05/2021 – 11/2021

Abstract:
Worldwide, Parkinson’s disease is the second most common degenerative disease with a prevalence of approximately 1% in people aged 60 years or older and causes increasing impairment of gait as the disease progresses [1]. Since the cause of Parkinson’s is still researched upon, only symptomatic treatment options such as drug-based dopamine replacement therapies or deep brain stimulation  are available [2]. As the disease progresses, patients need higher and more frequent medication intake, which, however, can lead to unwanted side effects such as bradykinesia as well as dyskinesia [3, 4]. In order to be able to guarantee optimal treatment, clinicians need precise information about the effect of medication on symptom occurrence and severity. However, since physicians adjust the medication based on their observation during irregular appointments with the patients in the laboratory, this precise information is not available to them. As several studies have shown significant differences in gait performance between laboratory and real-world conditions due to environmental effects and the active focus on walking [5, 6], real-world walking analysis may be a feasible way of  assessing actual symptoms. Furthermore, disease related symptoms fluctuate over time and should therefore be assessed on a continuous basis.

In order to enable long-term monitoring of the patient’s gait, inertial measurement units (IMUs) are often used, which are slightly less accurate than gold standard methods such as motion capture systems but are suitable for analyzing gait parameters such as stride length or stride time in continuous real world monitoring applications [6–8]. As the disease progresses, patients find it increasingly difficult to perform motor skills such as walking precisely, which may be reflected by a reduced stride length and greater stride variability [6, 9]. Since studies have already shown that the regular use of dopamine replacement drugs leads to fluctuations in motor skills throughout the day [9], macro/quantitative parameters (such as physical activity) and Parkinson’s-specific symptoms need also to be examined. Prominent Parkinson symptoms are resting tremor [10] and freezing of gait (FoG) [11]. There are already several approaches based on threshold and machine leaning for extracting the features from sensor data [8, 12] in order to detect and quantify those symptoms. However, a comprehensive analysis of the effects of medication on all qualitative and quantitative gait parameters in connection with Parkinson’s-specific symptoms does not yet exist in the literature.

In contrast to other studies, the aim of this study is to investigate whether the expression of the proposed parameters statistically correlate with the time of Levodopa intake and which micro/qualitative gait parameters are best suited to quantify medication effects in Parkinson’s treatment. For this purpose, sensor data from the FallRiskPD dataset will be used, which consists of two-week movement recordings of Parkinson’s patients (recorded with two IMUs on the ankles, one on the L5 posterior trunk) and the patient’s medication plan.

References:
[1] E. R. Dorsey et al., “Global, regional, and national burden of Parkinson’s disease, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016,” The Lancet Neurology, vol. 17, no. 11, pp. 939–953, 2018, doi: 10.1016/S1474-4422(18)30295-3.
[2] M. J. Armstrong and M. S. Okun, “Diagnosis and Treatment of Parkinson Disease: A Review,” JAMA, vol. 323, no. 6, pp. 548–560, 2020, doi: 10.1001/jama.2019.22360.
[3] B. S. Connolly and A. E. Lang, “Pharmacological treatment of Parkinson disease: a review,” JAMA, vol. 311, no. 16, pp. 1670–1683, 2014, doi: 10.1001/jama.2014.3654.
[4] S. Fahn et al., “Levodopa and the progression of Parkinson’s disease,” The New England journal of medicine, vol. 351, no. 24, pp. 2498–2508, 2004, doi: 10.1056/nejmoa033447.
[5] S. Del Din, A. Godfrey, B. Galna, S. Lord, and L. Rochester, “Free-living gait characteristics in ageing and Parkinson’s disease: impact of environment and ambulatory bout length,” J NeuroEngineering Rehabil, vol. 13, no. 1, p. 46, 2016, doi: 10.1186/s12984-016-0154-5.
[6] A. Atrsaei et al., “Gait speed in clinical and daily living assessments in Parkinson’s disease patients: performance versus capacity,” npj Parkinsons Dis., vol. 7, no. 1, p. 24, 2021, doi: 10.1038/s41531-021-00171-0.
[7] C. L. Pulliam, D. A. Heldman, E. B. Brokaw, T. O. Mera, Z. K. Mari, and M. A. Burack, “Continuous Assessment of Levodopa Response in Parkinson’s Disease Using Wearable Motion Sensors,” IEEE transactions on bio-medical engineering, vol. 65, no. 1, pp. 159–164, 2018, doi: 10.1109/tbme.2017.2697764.
[8] J. E. Thorp, P. G. Adamczyk, H.-L. Ploeg, and K. A. Pickett, “Monitoring Motor Symptoms During Activities of Daily Living in Individuals With Parkinson’s Disease,” Front. Neurol., vol. 9, p. 1036, 2018, doi: 10.3389/fneur.2018.01036.
[9] S. T. Moore, H. G. MacDougall, J.-M. Gracies, H. S. Cohen, and W. G. Ondo, “Long-term monitoring of gait in Parkinson’s disease,” Gait & Posture, vol. 26, no. 2, pp. 200–207, 2007, doi: 10.1016/j.gaitpost.2006.09.011.
[10] A. Papadopoulos, K. Kyritsis, L. Klingelhoefer, S. Bostanjopoulou, K. R. Chaudhuri, and A. Delopoulos, “Detecting Parkinsonian Tremor From IMU Data Collected in-the-Wild Using Deep Multiple-Instance Learning,” IEEE J. Biomed. Health Inform., vol. 24, no. 9, pp. 2559–2569, 2020, doi: 10.1109/jbhi.2019.2961748.
[11] M. Mancini et al., “Measuring freezing of gait during daily-life: an open-source, wearable sensors approach,” J NeuroEngineering Rehabil, vol. 18, no. 1, p. 1, 2021, doi: 10.1186/s12984-020-00774-3.
[12] R. San-Segundo et al., “Parkinson’s Disease Tremor Detection in the Wild Using Wearable Accelerometers,” Sensors (Basel, Switzerland), vol. 20, no. 20, p. 5817, 2020, doi: 10.3390/s20205817.