Detecting and monitoring the symptoms of Parkinson's disease using smartphones: A pilot study
Open accessAbstract
Background Remote, non-invasive and objective tests that can be used to support expert diagnosis for Parkinson's disease (PD) are lacking.Methods Participants underwent baseline in-clinic assessments, including the Unified Parkinson's Disease Rating Scale (UPDRS), and were provided smartphones with an Android operating system that contained a smartphone application that assessed voice, posture, gait, finger tapping, and response time. Participants then took the smart phones home to perform the five tasks four times a day for a month. Once a week participants had a remote (telemedicine) visit with a Parkinson disease specialist in which a modified (excluding assessments of rigidity and balance) UPDRS performed. Using statistical analyses of the five tasks recorded using the smartphone from 10 individuals with PD and 10 controls, we sought to: (1) discriminate whether the participant had PD and (2) predict the modified motor portion of the UPDRS.Results Twenty participants performed an average of 2.7 tests per day (68.9% adherence) for the study duration (average of 34.4 days) in a home and community setting. The analyses of the five tasks differed between those with Parkinson disease and those without. In discriminating participants with PD from controls, the mean sensitivity was 96.2% (SD 2%) and mean specificity was 96.9% (SD 1.9%). The mean error in predicting the modified motor component of the UPDRS (range 11-34) was 1.26 UPDRS points (SD 0.16).Conclusion Measuring PD symptoms via a smartphone is feasible and has potential value as a diagnostic support tool.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Using machine learning algorithms to map wearable sensor data onto existing clinical scales such as UPDRS is a poor approach because wearable sensors can be far more sensitive and precise than subjective ratings.
“The specific clinimetric validation approach taken needs some care: although it is possible to use machine learning algorithms to map wearable sensor data onto existing clinical scales such as UPDRS 9 , this is a poor approach because wearable sensors can be far more sensitive and precise than subjective ratings.”
From Machine learning for large‐scale wearable sensor data in Parkinson's disease: Concepts, promises, pitfalls, and futures (2016) · cites this paper as [9]
The Results report that a random-forest model predicted modified motor UPDRS scores from smartphone measurements, supporting the cited example of mapping sensor data to a clinical scale. The Introduction distinguishes smartphones from wearable sensors; this study does not establish the wearable-sensor example specifically. The criticism after citation 9 is the citing authors’ own assessment, not a point attributed to this paper. (LLM, conf=0.82)
Experimental — accuracy not yet measuredRegression has been used to predict UPDRS ratings from digital voice recordings, accelerometers, and touch screens.
“This approach has been used to predict UPDRS ratings from digital voice recordings, accelerometers, and touch screens. 9”
From Machine learning for large‐scale wearable sensor data in Parkinson's disease: Concepts, promises, pitfalls, and futures (2016) · cites this paper as [9]
Arora et al. used random forests in discriminating patients with PD from controls from a broad variety of smartphone sensor data.
“Arora et al. 9 used random forests in discriminating patients with PD from controls from a broad variety of smartphone sensor data.”
From Machine learning for large‐scale wearable sensor data in Parkinson's disease: Concepts, promises, pitfalls, and futures (2016) · cites this paper as [9]
Arora et al. (2015) reported an average mean absolute error of 1.26 points in modified motor UPDRS score.
“average mean absolute error in modified motor UPDRS score of 1.26 points”
From Machine learning for large‐scale wearable sensor data in Parkinson's disease: Concepts, promises, pitfalls, and futures (2016) · cites this paper as [9]
Arora et al. (2015) reported that a random forest classifier using voice recording and smartphone touchscreen data achieved a sensitivity of 96.2% and specificity of 96.9% in PD diagnosis.
“sensitivity of 96.2% and specificity of 96.9% in PD diagnosis”
From Machine learning for large‐scale wearable sensor data in Parkinson's disease: Concepts, promises, pitfalls, and futures (2016) · cites this paper as [9]
Arora and colleagues deployed an Android smartphone to 10 individuals with PD and 10 controls who performed four active tests four times daily for 1 month; adherence was acceptable at 69%, and random forest machine-learning successfully discriminated PD participants from controls.
“Arora and colleagues19 deployed an Android smartphone to 10 individuals with PD (mean motor MDS‐UPDRS score = 19.6 [standard deviation {SD} = 6.7]) and 10 controls, who performed four active tests (sustained phonation, gait, finger‐tapping, and reaction time) four times daily for 1 month. Adherence was acceptable at 69%, and random forest machine‐learning successfully discriminated PD participants”
From Evaluation of smartphone‐based testing to generate exploratory outcome measures in a phase 1 Parkinson's disease clinical trial (2018) · cites this paper as [19]
Adherence was similar to a smaller-scale digital biomarker approach with four active tests performed four times daily for 1 month (69%).
“similar to a smaller‐scale digital biomarker approach with four active tests performed four times daily for 1 month (69%19)”
From Evaluation of smartphone‐based testing to generate exploratory outcome measures in a phase 1 Parkinson's disease clinical trial (2018) · cites this paper as [19]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 10 · Cited by 387
- Authors
- Arora, S., Venkataraman, V., Zhan, A., Donohue, S., Biglan, K. M., Dorsey, E. R., & Little, M. A.
- Affiliations
- Aston University (GB)
- Duke University (US)
- Johns Hopkins University (US)
- University of Rochester Medicine (US)
- Corresponding author
- Dorsey, E.R.
- Venue
- Parkinsonism & Related Disorders
- Publisher
- Elsevier BV
- Published
- 2015-06
- Volume
- 21
- Issue
- 6
- Pages
- 650-653
- ISSN
- 1353-8020 (ISSN-L)
- OpenAlex
- W2095134828
- Cited by
- 387
- References
- 10
- Citations by year
- 2026: 27 · 2025: 24 · 2024: 27 · 2023: 28 · 2022: 37 · 2021: 41 · 2020: 42 · 2019: 51 · 2018: 35 · 2017: 36 · 2016: 30 · 2015: 9
- FWCI
- 25.84
- Type
- journal article
- Language
- en
- Open access
- Locations
- Parkinsonism & Related Disorders (publishedVersion)
- PubMed (publishedVersion)
- Aston Publications Explorer (Aston University) (submittedVersion)
- Funders
- Patient-Centered Outcomes Research Institute
- Huntington's Disease Society of America
- Davis Phinney Foundation
- National Parkinson Foundation
- Wellcome Trust
- National Institute of Neurological Disorders and Stroke
- UCB
- Michael J. Fox Foundation for Parkinson's Research
- Great Lakes Neurotechnologies
- Medtronic
- H. Lundbeck A/S
- Topics
- Health Sciences › Medicine › Neurology › Parkinson's Disease Mechanisms and Treatments
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Health Professions › Physical Therapy, Sports Therapy and Rehabilitation › Balance, Gait, and Falls Prevention
- Keywords
- Parkinson's disease, smartphone-based assessment
- MeSH terms
- Smartphone, Aged, Female, Fingers, Gait, Humans, Male, Middle Aged, Motor Skills, Parkinson Disease, Pilot Projects, Posture, Reaction Time, Sensitivity and Specificity, Severity of Illness Index, Voice, Telemedicine
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
