Dataset Information

Linear predictive coding distinguishes spectral EEG features of Parkinson's disease.

ABSTRACT:

Objective

We have developed and validated a novel EEG-based signal processing approach to distinguish PD and control patients: Linear-predictive-coding EEG Algorithm for PD (LEAPD). This method efficiently encodes EEG time series into features that can detect PD in a computationally fast manner amenable to real time applications.

Methods

We included a total of 41 PD patients and 41 demographically-matched controls from New Mexico and Iowa. Data for all participants from New Mexico (27 PD patients and 27 controls) were used to evaluate in-sample LEAPD performance, with extensive cross-validation. Participants from Iowa (14 PD patients and 14 controls) were used for out-of-sample tests. Our method utilized data from six EEG leads which were as little as 2 min long.

Results

For the in-sample dataset, LEAPD differentiated PD patients from controls with 85.3 ± 0.1% diagnostic accuracy, 93.3 ± 0.5% area under the receiver operating characteristics curve (AUC), 87.9 ± 0.9% sensitivity, and 82.7 ± 1.1% specificity, with multiple cross-validations. After head-to-head comparison with state-of-the-art methods using our dataset, LEAPD showed a 13% increase in accuracy and a 15.5% increase in AUC. When the trained classifier was applied to a distinct out-of-sample dataset, LEAPD showed reliable performance with 85.7% diagnostic accuracy, 85.2% AUC, 85.7% sensitivity, and 85.7% specificity. No statistically significant effect of levodopa-ON and levodopa-OFF sessions were found.

Conclusion

We describe LEAPD, an efficient algorithm that is suitable for real time application and captures spectral EEG features using few parameters and reliably differentiates PD patients from demographically-matched controls.

SUBMITTER: Anjum MF

PROVIDER: S-EPMC7900258 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Repetitive finger tapping is commonly used to assess bradykinesia in Parkinson's disease. The Queen Square Brain Bank diagnostic criterion of Parkinson's disease defines bradykinesia as 'slowness of initiation with progressive reduction in speed and amplitude of repetitive action'. Although progressive supranuclear palsy is considered an atypical parkinsonian syndrome, it is not known whether patients with progressive supranuclear palsy have criteria-defined bradykinesia. This study objectively assessed repetitive finger tap performance and handwriting in patients with Parkinson's disease (n?=?15), progressive supranuclear palsy (n?=?9) and healthy age- and gender-matched controls (n?=?16). The motion of the hand and digits was recorded in 3D during 15-s repetitive index finger-to-thumb tapping trials. The main finding was hypokinesia without decrement in patients with progressive supranuclear palsy, which differed from the finger tap pattern in Parkinson's disease. Average finger separation amplitude in progressive supranuclear palsy was less than half of that in controls and Parkinson's disease (P?<?0.001 in both cases). Change in tap amplitude over consecutive taps was computed by linear regression. The average amplitude slope in progressive supranuclear palsy was nearly zero (0.01°/cycle) indicating a lack of decrement, which differed from the negative slope in patients with Parkinson's disease OFF levodopa (-0.20°/cycle, P?=?0.002). 'Hypokinesia', defined as <50% of control group's mean amplitude, combined with 'absence of decrement', defined as mean positive amplitude slope, were identified in 87% of finger tap trials in the progressive supranuclear palsy group and only 12% in the Parkinson's disease OFF levodopa group. In progressive supranuclear palsy, the mean amplitude was not correlated with disease duration or other clinimetric scores. In Parkinson's disease, finger tap pattern was compatible with criteria-defined bradykinesia, characterized by slowness with progressive reduction in amplitude and speed and increased variability in speed throughout the tap trial. In Parkinson's disease, smaller amplitude, slower speed and greater speed variability were all associated with a more severe Unified Parkinson's Disease Rating Scale motor score. Analyses of handwriting showed that micrographia, defined as smaller than 50% of the control group's mean script size, was present in 75% of patients with progressive supranuclear palsy and 15% of patients with Parkinson's disease (P?=?0.022). Most scripts performed by patients with progressive supranuclear palsy did not exhibit decrements in script size. In conclusion, patients with progressive supranuclear palsy have a specific finger tap pattern of 'hypokinesia without decrement' and they do not have criteria-defined limb bradykinesia. Similarly, 'micrographia' and 'lack of decrement in script size' are also more common in progressive supranuclear palsy than in Parkinson's disease.