Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach
Abstract Background Postural instability greatly reduces quality of life in people with Parkinson’s disease (PD). Early and objective detection of postural impairments is crucial to facilitate interventions. Our aim was to use a convolutional neural network (CNN) to differentiate people with early t...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | Journal of NeuroEngineering and Rehabilitation |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12984-025-01570-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850190530113175552 |
|---|---|
| author | David Engel R. Stefan Greulich Alberto Parola Kaleb Vinehout Justus Student Josefine Waldthaler Lars Timmermann Frank Bremmer |
| author_facet | David Engel R. Stefan Greulich Alberto Parola Kaleb Vinehout Justus Student Josefine Waldthaler Lars Timmermann Frank Bremmer |
| author_sort | David Engel |
| collection | DOAJ |
| description | Abstract Background Postural instability greatly reduces quality of life in people with Parkinson’s disease (PD). Early and objective detection of postural impairments is crucial to facilitate interventions. Our aim was to use a convolutional neural network (CNN) to differentiate people with early to mid-stage PD from healthy age-matched individuals based on spectrogram images obtained from their body sway. We hypothesized the time–frequency content of body sway to be predictive of PD, even when impairments are not yet clinically apparent. Methods 18 people with idiopathic PD and 15 healthy controls (HC) participated in the study. We tracked participants’ center of pressure (COP) using a Wii Balance Board and their full-body motion using a Microsoft Kinect, out of which we calculated the trajectory of their center of mass (COM). We used 30 s-snippets of motion data from which we acquired wavelet-based time–frequency spectrograms that were fed into a custom-built CNN as labeled images. We used binary classification to have the network differentiate between individuals with PD and controls (n = 15, respectively). Results Classification performance was best when the medio-lateral motion of the COM was considered. Here, our network reached a predictive accuracy, sensitivity, specificity, precision and F1-score of 100%, respectively, with a receiver operating characteristic area under the curve of 1.0. Moreover, an explainable AI approach revealed high frequencies in the postural sway data to be most distinct between both groups. Conclusion Heeding our small and heterogeneous sample, our findings suggest a CNN classifier based on cost-effective and conveniently obtainable posturographic data to be a promising approach to detect postural impairments in early to mid-stage PD and to gain novel insight into the subtle characteristics of impairments at this stage of the disease. |
| format | Article |
| id | doaj-art-e2d7cc3ef76d436b962c91d73c88a520 |
| institution | OA Journals |
| issn | 1743-0003 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Journal of NeuroEngineering and Rehabilitation |
| spelling | doaj-art-e2d7cc3ef76d436b962c91d73c88a5202025-08-20T02:15:15ZengBMCJournal of NeuroEngineering and Rehabilitation1743-00032025-02-0122111310.1186/s12984-025-01570-7Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approachDavid Engel0R. Stefan Greulich1Alberto Parola2Kaleb Vinehout3Justus Student4Josefine Waldthaler5Lars Timmermann6Frank Bremmer7Applied Physics and Neurophysics, Philipps-Universität MarburgChair of Business Information Systems, Esp. Intelligent Systems and Services, TUD Dresden University of TechnologyCentre for Language Technology, Department of Nordic Studies and Linguistics , Copenhagen UniversityCold Spring Harbor Laboratory (CSHL)Department of Neurology, University Hospital Giessen and MarburgCenter for Mind, Brain and Behavior (CMBB), Philipps-Universität MarburgCenter for Mind, Brain and Behavior (CMBB), Philipps-Universität MarburgApplied Physics and Neurophysics, Philipps-Universität MarburgAbstract Background Postural instability greatly reduces quality of life in people with Parkinson’s disease (PD). Early and objective detection of postural impairments is crucial to facilitate interventions. Our aim was to use a convolutional neural network (CNN) to differentiate people with early to mid-stage PD from healthy age-matched individuals based on spectrogram images obtained from their body sway. We hypothesized the time–frequency content of body sway to be predictive of PD, even when impairments are not yet clinically apparent. Methods 18 people with idiopathic PD and 15 healthy controls (HC) participated in the study. We tracked participants’ center of pressure (COP) using a Wii Balance Board and their full-body motion using a Microsoft Kinect, out of which we calculated the trajectory of their center of mass (COM). We used 30 s-snippets of motion data from which we acquired wavelet-based time–frequency spectrograms that were fed into a custom-built CNN as labeled images. We used binary classification to have the network differentiate between individuals with PD and controls (n = 15, respectively). Results Classification performance was best when the medio-lateral motion of the COM was considered. Here, our network reached a predictive accuracy, sensitivity, specificity, precision and F1-score of 100%, respectively, with a receiver operating characteristic area under the curve of 1.0. Moreover, an explainable AI approach revealed high frequencies in the postural sway data to be most distinct between both groups. Conclusion Heeding our small and heterogeneous sample, our findings suggest a CNN classifier based on cost-effective and conveniently obtainable posturographic data to be a promising approach to detect postural impairments in early to mid-stage PD and to gain novel insight into the subtle characteristics of impairments at this stage of the disease.https://doi.org/10.1186/s12984-025-01570-7Parkinson’s diseasePostural instabilityBody swayCenter of pressureCenter of massFrequency analysis |
| spellingShingle | David Engel R. Stefan Greulich Alberto Parola Kaleb Vinehout Justus Student Josefine Waldthaler Lars Timmermann Frank Bremmer Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach Journal of NeuroEngineering and Rehabilitation Parkinson’s disease Postural instability Body sway Center of pressure Center of mass Frequency analysis |
| title | Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach |
| title_full | Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach |
| title_fullStr | Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach |
| title_full_unstemmed | Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach |
| title_short | Sway frequencies may predict postural instability in Parkinson’s disease: a novel convolutional neural network approach |
| title_sort | sway frequencies may predict postural instability in parkinson s disease a novel convolutional neural network approach |
| topic | Parkinson’s disease Postural instability Body sway Center of pressure Center of mass Frequency analysis |
| url | https://doi.org/10.1186/s12984-025-01570-7 |
| work_keys_str_mv | AT davidengel swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT rstefangreulich swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT albertoparola swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT kalebvinehout swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT justusstudent swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT josefinewaldthaler swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT larstimmermann swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach AT frankbremmer swayfrequenciesmaypredictposturalinstabilityinparkinsonsdiseaseanovelconvolutionalneuralnetworkapproach |