Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network
Deep convolutional neural network (DCNN) provides a multivariate framework to detect relevant spatio-oscillatory patterns in the data beyond common mass-univariate statistics. Yet, its practical application is limited due to the low interpretability of the results beyond accuracy. We opted to use DC...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-01-01
|
| Series: | NeuroImage: Clinical |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S221315822200331X |
| _version_ | 1811224634266222592 |
|---|---|
| author | Mehran Shabanpour Neda Kaboodvand Behzad Iravani |
| author_facet | Mehran Shabanpour Neda Kaboodvand Behzad Iravani |
| author_sort | Mehran Shabanpour |
| collection | DOAJ |
| description | Deep convolutional neural network (DCNN) provides a multivariate framework to detect relevant spatio-oscillatory patterns in the data beyond common mass-univariate statistics. Yet, its practical application is limited due to the low interpretability of the results beyond accuracy. We opted to use DCNN with a minimalistic architecture design and large penalized terms to yield a generalizable and clinically relevant network model. Our network was trained based on the scalp topology of the electroencephalography (EEG) from an open access dataset, constituting our primary sample of healthy controls (n = 25) and Parkinson’s disease (PD) patients (n = 25), with and without medication. Next, we validated the model on another independent, yet comparable open access EEG dataset (healthy controls (n = 20) and PD patients (n = 20)), which was unseen to the network. We applied Gradient-weighted Class Activation Mapping (Grad-CAM) interpretability technique to create a localization map exhibiting the key network predictors, based on the gradients of the classification score flowing into the last convolutional layer. Accordingly, our results indicated that a sub-second of intrinsic oscillatory power pattern in the beta band over the occipitoparietal, gamma band over the left motor cortex as well as theta band over the frontoparietal cluster, had the largest impact on the network score for dissociating the PD patients from age- and gender-matched healthy controls, across the two datasets. We further found that the off-medication motor symptoms were related to the occipitoparietal off-medication beta power whereas the disease duration was associated with the off-medication beta power of the motor cortex. The on-medication theta power of the frontoparietal was related to the improvement of the motor symptoms. In conclusion, our method enabled us to characterize PD patho-electrophysiology according to the multivariate topographic analysis approach, where both spatial and frequency aspects of the oscillations were simultaneously considered. Moreover, our approach was free from common reference problem of the EEG data analyses. |
| first_indexed | 2024-04-12T08:52:25Z |
| format | Article |
| id | doaj.art-0fb64f60284c4a299be78fcc323a6479 |
| institution | Directory Open Access Journal |
| issn | 2213-1582 |
| language | English |
| last_indexed | 2024-04-12T08:52:25Z |
| publishDate | 2022-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage: Clinical |
| spelling | doaj.art-0fb64f60284c4a299be78fcc323a64792022-12-22T03:39:32ZengElsevierNeuroImage: Clinical2213-15822022-01-0136103266Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural networkMehran Shabanpour0Neda Kaboodvand1Behzad Iravani2Zanjan University of Medical Sciences, Zanjan, IranDepartment of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; Department of Neurology and Neurological Science, Stanford University, Stanford, United StatesDepartment of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; Department of Neurology and Neurological Science, Stanford University, Stanford, United States; Corresponding author at: Full postal address: K8 Klinisk neurovetenskap, K8 Neuro Fransson, 171 77 Stockholm, Sweden.Deep convolutional neural network (DCNN) provides a multivariate framework to detect relevant spatio-oscillatory patterns in the data beyond common mass-univariate statistics. Yet, its practical application is limited due to the low interpretability of the results beyond accuracy. We opted to use DCNN with a minimalistic architecture design and large penalized terms to yield a generalizable and clinically relevant network model. Our network was trained based on the scalp topology of the electroencephalography (EEG) from an open access dataset, constituting our primary sample of healthy controls (n = 25) and Parkinson’s disease (PD) patients (n = 25), with and without medication. Next, we validated the model on another independent, yet comparable open access EEG dataset (healthy controls (n = 20) and PD patients (n = 20)), which was unseen to the network. We applied Gradient-weighted Class Activation Mapping (Grad-CAM) interpretability technique to create a localization map exhibiting the key network predictors, based on the gradients of the classification score flowing into the last convolutional layer. Accordingly, our results indicated that a sub-second of intrinsic oscillatory power pattern in the beta band over the occipitoparietal, gamma band over the left motor cortex as well as theta band over the frontoparietal cluster, had the largest impact on the network score for dissociating the PD patients from age- and gender-matched healthy controls, across the two datasets. We further found that the off-medication motor symptoms were related to the occipitoparietal off-medication beta power whereas the disease duration was associated with the off-medication beta power of the motor cortex. The on-medication theta power of the frontoparietal was related to the improvement of the motor symptoms. In conclusion, our method enabled us to characterize PD patho-electrophysiology according to the multivariate topographic analysis approach, where both spatial and frequency aspects of the oscillations were simultaneously considered. Moreover, our approach was free from common reference problem of the EEG data analyses.http://www.sciencedirect.com/science/article/pii/S221315822200331XConvolutional neural networkResting-state oscillationParkinson’s diseaseMotor cortical beta activityFrontoparietal theta power |
| spellingShingle | Mehran Shabanpour Neda Kaboodvand Behzad Iravani Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network NeuroImage: Clinical Convolutional neural network Resting-state oscillation Parkinson’s disease Motor cortical beta activity Frontoparietal theta power |
| title | Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network |
| title_full | Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network |
| title_fullStr | Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network |
| title_full_unstemmed | Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network |
| title_short | Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network |
| title_sort | parkinson s disease is characterized by sub second resting state spatio oscillatory patterns a contribution from deep convolutional neural network |
| topic | Convolutional neural network Resting-state oscillation Parkinson’s disease Motor cortical beta activity Frontoparietal theta power |
| url | http://www.sciencedirect.com/science/article/pii/S221315822200331X |
| work_keys_str_mv | AT mehranshabanpour parkinsonsdiseaseischaracterizedbysubsecondrestingstatespatiooscillatorypatternsacontributionfromdeepconvolutionalneuralnetwork AT nedakaboodvand parkinsonsdiseaseischaracterizedbysubsecondrestingstatespatiooscillatorypatternsacontributionfromdeepconvolutionalneuralnetwork AT behzadiravani parkinsonsdiseaseischaracterizedbysubsecondrestingstatespatiooscillatorypatternsacontributionfromdeepconvolutionalneuralnetwork |