Journal article
Linear predictive coding electroencephalography algorithms predict mortality in Parkinson’s disease
Clinical parkinsonism & related disorders, Vol.13, 100409
2025
DOI: 10.1016/j.prdoa.2025.100409
PMCID: PMC12719220
PMID: 41439144
Abstract
•Mortality increases in Parkinson's disease (PD) and is difficult to predict because of lack of reliable prognostic markers.•We used two minutes of EEG recordings and the Linear Predictive Coding EEG Algorithm for PD (LEAPD) to assess mortality.•We used LEAPD to classify 3-year mortality status and correlate LEAPD indices with time to death.•Out of sample tests gave high classification accuracy and high Spearman’s correlation between LEAPD indices and time to death.•Short resting state EEG using LEAPD can predict mortality and time to death in PD.
Mortality is increased in Parkinson’s disease (PD) and is difficult to predict because of its heterogeneity and the availability of few reliable prognostic markers.
We used electroencephalography (EEG) recordings and the Linear Predictive Coding EEG Algorithm for PD (LEAPD) to classify 3-year mortality status and correlate LEAPD indices with time to death.
2-minutes resting-state EEG from 94 PD patients was used for binary classification of 3-year mortality status (22 deceased). Single-channel classification using a balanced dataset of 44 was performed using leave-one-out cross-validation (LOOCV). Robustness was evaluated by truncating the recordings. LOOCV Spearman’s correlation coefficient (ρ) was obtained between LEAPD indices and time to death. Optimum hyperparameters obtained using a balanced training dataset of 30 were tested on the remaining 64 by 10,000 randomized comparisons of 7 vs 7, using 5 channel combinations. Separate hyperparameters for the best ρ, obtained using the same training dataset, were used for out-of-sample correlation for the remaining 7 deceased.
The deceased participants were older and had more severe disease and worse cognition at baseline. Several EEG channels yielded 100 % LOOCV accuracy. Five had robust performance under data truncation. The correlations ranged between ρ = −0.59 to −0.86 and were significant after adjusting for age, cognitive and motor impairment. Out-of-sample testing using the best-performing 5-channel combination yielded a mean accuracy of 83 %. Out-of-sample Spearman’s ρ was −0.82.
Short resting EEG using machine learning algorithms such as linear predictive coding can predict mortality in PD.
Details
- Title: Subtitle
- Linear predictive coding electroencephalography algorithms predict mortality in Parkinson’s disease
- Creators
- Simin Jamshidi - University of IowaArturo I. Espinoza - University of IowaJonathan T. Heinzman - Broadlawns Medical CenterPatrick May - Westford University CollegeErgun Y. Uc - University of IowaNandakumar S. Narayanan - University of IowaSoura Dasgupta - Department of Computer and Electrical Engineering, College of Engineering, University of Iowa, Iowa City, IA, USA
- Resource Type
- Journal article
- Publication Details
- Clinical parkinsonism & related disorders, Vol.13, 100409
- DOI
- 10.1016/j.prdoa.2025.100409
- PMID
- 41439144
- PMCID
- PMC12719220
- NLM abbreviation
- Clin Park Relat Disord
- ISSN
- 2590-1125
- eISSN
- 2590-1125
- Publisher
- Elsevier Ltd
- Grant note
- National Institutes of Health: R01NS100849-01A1, RF1NS127809-01A1 National Institutes of Health (NIH)
This study was supported by the National Institutes of Health under grants R01NS100849-01A1 and RF1NS127809-01A1. The National Institute of Neurological Disorders and Stroke (NINDS) was not involved in the study's design, data collection, analysis, interpretation, or manuscript preparation. We sincerely thank the staff of the Department of Neurology at the University of Iowa for their support and assistance throughout this research. We thank Heather Widmayer of the Scientific Editing and Research Communication Core at the University of Iowa Carver College of Med-icine for her editorial assistance. We thank Dr, Md Fahim Anjum for assistance with the LEAPD algorithm. This manuscript is the result of funding in whole or in part by the National Institutes of Health (NIH) . It is subject to the NIH Public Access Policy. Through acceptance of this federal funding, NIH has been given a right to make this manuscript publicly available in PubMed Central upon the Official Date of Publication, as defined by NIH.
- Language
- English
- Date published
- 2025
- Academic Unit
- Neurology; Electrical and Computer Engineering; Psychiatry; Iowa Neuroscience Institute
- Record Identifier
- 9985090635302771
Metrics
6 Record Views