Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering
Objective Accurate and reliable detection of tremor onset in Parkinson’s disease (PD) is critical to the success of adaptive deep brain stimulation (aDBS) therapy. Here, we investigated the potential use of feature engineering and machine learning methods for more accurate detection of rest tremor i...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Elsevier
2019
|
| _version_ | 1797068865377665024 |
|---|---|
| author | Yao, L Brown, P Shoaran, M |
| author_facet | Yao, L Brown, P Shoaran, M |
| author_sort | Yao, L |
| collection | OXFORD |
| description | Objective
Accurate and reliable detection of tremor onset in Parkinson’s disease (PD) is critical to the success of adaptive deep brain stimulation (aDBS) therapy. Here, we investigated the potential use of feature engineering and machine learning methods for more accurate detection of rest tremor in PD.
<br></br>
Methods
We analyzed the local field potential (LFP) recordings from the subthalamic nucleus region in 12 patients with PD (16 recordings). To explore the optimal biomarkers and the best performing classifier, the performance of state-of-the-art machine learning (ML) algorithms and various features of the subthalamic LFPs were compared. We further used a Kalman filtering technique in feature domain to reduce the false positive rate.
<br></br>
Results
The Hjorth complexity showed a higher correlation with tremor, compared to other features in our study. In addition, by optimal selection of a maximum of five features with a sequential feature selection method and using the gradient boosted decision trees as the classifier, the system could achieve an average F1 score of up to 88.7% and a detection lead of 0.52 s. The use of Kalman filtering in feature space significantly improved the specificity by 17.0% (p = 0.002), thereby potentially reducing the unnecessary power dissipation of the conventional DBS system.
<br></br>
Conclusion
The use of relevant features combined with Kalman filtering and machine learning improves the accuracy of tremor detection during rest.
<br></br>
Significance
The proposed method offers a potential solution for efficient on-demand stimulation for PD tremor. |
| first_indexed | 2024-03-06T22:16:08Z |
| format | Journal article |
| id | oxford-uuid:53727cfc-ff8d-4c4c-a6c2-de01a82b3147 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T22:16:08Z |
| publishDate | 2019 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:53727cfc-ff8d-4c4c-a6c2-de01a82b31472022-03-26T16:31:41ZImproved detection of Parkinsonian resting tremor with feature engineering and Kalman filteringJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:53727cfc-ff8d-4c4c-a6c2-de01a82b3147EnglishSymplectic ElementsElsevier 2019Yao, LBrown, PShoaran, MObjective Accurate and reliable detection of tremor onset in Parkinson’s disease (PD) is critical to the success of adaptive deep brain stimulation (aDBS) therapy. Here, we investigated the potential use of feature engineering and machine learning methods for more accurate detection of rest tremor in PD. <br></br> Methods We analyzed the local field potential (LFP) recordings from the subthalamic nucleus region in 12 patients with PD (16 recordings). To explore the optimal biomarkers and the best performing classifier, the performance of state-of-the-art machine learning (ML) algorithms and various features of the subthalamic LFPs were compared. We further used a Kalman filtering technique in feature domain to reduce the false positive rate. <br></br> Results The Hjorth complexity showed a higher correlation with tremor, compared to other features in our study. In addition, by optimal selection of a maximum of five features with a sequential feature selection method and using the gradient boosted decision trees as the classifier, the system could achieve an average F1 score of up to 88.7% and a detection lead of 0.52 s. The use of Kalman filtering in feature space significantly improved the specificity by 17.0% (p = 0.002), thereby potentially reducing the unnecessary power dissipation of the conventional DBS system. <br></br> Conclusion The use of relevant features combined with Kalman filtering and machine learning improves the accuracy of tremor detection during rest. <br></br> Significance The proposed method offers a potential solution for efficient on-demand stimulation for PD tremor. |
| spellingShingle | Yao, L Brown, P Shoaran, M Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title | Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title_full | Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title_fullStr | Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title_full_unstemmed | Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title_short | Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering |
| title_sort | improved detection of parkinsonian resting tremor with feature engineering and kalman filtering |
| work_keys_str_mv | AT yaol improveddetectionofparkinsonianrestingtremorwithfeatureengineeringandkalmanfiltering AT brownp improveddetectionofparkinsonianrestingtremorwithfeatureengineeringandkalmanfiltering AT shoaranm improveddetectionofparkinsonianrestingtremorwithfeatureengineeringandkalmanfiltering |