Benchmarking of EEG Analysis Techniques for Parkinson's Disease Diagnosis: A Comparison between Traditional ML Methods and Foundation DL Methods

2025-07-18

AI Summary

Existing studies on automated Parkinson’s disease (PD) diagnosis from electroencephalography (EEG) lack standardized preprocessing and reliable, subject-independent validation, hindering fair comparison between traditional machine learning and deep learning approaches. Method: We propose a standardized seven-step EEG preprocessing pipeline and a subject-independent cross-validation framework, enabling the first reproducible, apples-to-apples benchmark of conventional models (e.g., XGBoost) versus deep learning architectures (e.g., CNN-LSTM) on PD-EEG data acquired during an oddball paradigm, using event-related potential (ERP) features. Contribution/Results: Under identical data and evaluation protocols, CNN-LSTM achieves the highest overall diagnostic performance, while XGBoost attains competitive accuracy and superior decision calibration—jointly establishing a robust baseline for PD-EEG classification. This work delivers an open-source, standardized evaluation paradigm and a reproducible benchmark for EEG-based neurodiagnosis.

Technology Category

• Machine Learning: Deep Learning Algorithms
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Parkinson's Disease PD is a progressive neurodegenerative disorder that affects motor and cognitive functions with early diagnosis being critical for effective clinical intervention Electroencephalography EEG offers a noninvasive and costeffective means of detecting PDrelated neural alterations yet the development of reliable automated diagnostic models remains a challenge In this study we conduct a systematic benchmark of traditional machine learning ML and deep learning DL models for classifying PD using a publicly available oddball task dataset Our aim is to lay the groundwork for developing an effective learning system and to determine which approach produces the best results We implement a unified sevenstep preprocessing pipeline and apply consistent subjectwise crossvalidation and evaluation criteria to ensure comparability across models Our results demonstrate that while baseline deep learning architectures particularly CNNLSTM models achieve the best performance compared to other deep learning architectures underlining the importance of capturing longrange temporal dependencies several traditional classifiers such as XGBoost also offer strong predictive accuracy and calibrated decision boundaries By rigorously comparing these baselines our work provides a solid reference framework for future studies aiming to develop and evaluate more complex or specialized architectures Establishing a reliable set of baseline results is essential to contextualize improvements introduced by novel methods ensuring scientific rigor and reproducibility in the evolving field of EEGbased neurodiagnostics

Problem

• Compare traditional ML and DL methods for EEG-based Parkinson's diagnosis
• Evaluate performance of CNN-LSTM and XGBoost in PD classification
• Establish standardized benchmarks for EEG analysis in neurodiagnostics

Innovation

• Benchmarking traditional ML and DL methods
• Using CNN-LSTM for temporal dependencies
• Applying unified preprocessing and evaluation