An explainable machine learning framework for predicting driving states using electroencephalogram

IF 2.3 4区医学 Q3 ENGINEERING, BIOMEDICAL

Medical Engineering & Physics Pub Date : 2025-05-09 DOI:10.1016/j.medengphy.2025.104355

Iqram Hussain , Se-Jin Park , AKM Azad , Salem Ali Alyami

{"title":"An explainable machine learning framework for predicting driving states using electroencephalogram","authors":"Iqram Hussain , Se-Jin Park , AKM Azad , Salem Ali Alyami","doi":"10.1016/j.medengphy.2025.104355","DOIUrl":null,"url":null,"abstract":"<div><h3>Objectives</h3><div>Understanding drivers' cognitive load is essential for enhancing road safety, as cognitive demands fluctuate across different driving scenarios, potentially impacting performance, and safety, particularly for drivers with neurological disabilities. This study aims to predict driving states in healthy adult drivers using electroencephalogram (EEG) and machine learning (ML) models; and interpret the neural activity associated with each driving condition.</div></div><div><h3>Methods</h3><div>EEG data were collected from participants using a Cognionics Quick-20 EEG headset in Resting state, City-way, Highway, and Suburb-way driving states in 360 full-screen real car cabin inside the driving simulator. Participants drove while experiencing varied cognitive workloads due to various driving environments. EEG Features, including spectral band powers and power ratios, were extracted to extract neural activity patterns relevant to driving states. Gradient Boosting (GBoost) and Random Forest (RF) machine-learning classifiers were applied to classify the driving states based on EEG features. A model-agnostic explanation approach was implemented for model interpretability, which highlighted EEG spectral features contributing to each driving state.</div></div><div><h3>Results</h3><div>The GBoost model achieved the highest classification accuracy among tested models, with ROC-AUC values of 1.00 for Resting, 0.96 for Suburb-Way, 0.91 for Highway, and 0.90 for City-Way states, effectively distinguishing driving states using EEG features. Model agnostic feature attribution approach highlighted key EEG contributors, such as power in Theta and Delta bands, and power ratios (DAR, DTR), providing interpretability and aligning with established neurological indicators of cognitive workload.</div></div><div><h3>Conclusion</h3><div>This study demonstrates the potential of EEG-based features combined with interpretable machine learning for driver states prediction. The approach offers a foundation for personalized, responsive driving assistance systems at improving road safety and preventing disability by monitoring to drivers' cognitive states.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104355"},"PeriodicalIF":2.3000,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical Engineering & Physics","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1350453325000748","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Objectives

Understanding drivers' cognitive load is essential for enhancing road safety, as cognitive demands fluctuate across different driving scenarios, potentially impacting performance, and safety, particularly for drivers with neurological disabilities. This study aims to predict driving states in healthy adult drivers using electroencephalogram (EEG) and machine learning (ML) models; and interpret the neural activity associated with each driving condition.

Methods

EEG data were collected from participants using a Cognionics Quick-20 EEG headset in Resting state, City-way, Highway, and Suburb-way driving states in 360 full-screen real car cabin inside the driving simulator. Participants drove while experiencing varied cognitive workloads due to various driving environments. EEG Features, including spectral band powers and power ratios, were extracted to extract neural activity patterns relevant to driving states. Gradient Boosting (GBoost) and Random Forest (RF) machine-learning classifiers were applied to classify the driving states based on EEG features. A model-agnostic explanation approach was implemented for model interpretability, which highlighted EEG spectral features contributing to each driving state.

Results

The GBoost model achieved the highest classification accuracy among tested models, with ROC-AUC values of 1.00 for Resting, 0.96 for Suburb-Way, 0.91 for Highway, and 0.90 for City-Way states, effectively distinguishing driving states using EEG features. Model agnostic feature attribution approach highlighted key EEG contributors, such as power in Theta and Delta bands, and power ratios (DAR, DTR), providing interpretability and aligning with established neurological indicators of cognitive workload.

Conclusion

This study demonstrates the potential of EEG-based features combined with interpretable machine learning for driver states prediction. The approach offers a foundation for personalized, responsive driving assistance systems at improving road safety and preventing disability by monitoring to drivers' cognitive states.

查看原文本刊更多论文

一个可解释的机器学习框架，用于使用脑电图预测驾驶状态

了解驾驶员的认知负荷对于提高道路安全至关重要，因为认知需求在不同的驾驶场景中会波动，可能会影响驾驶表现和安全，特别是对于神经系统残疾的驾驶员。本研究旨在利用脑电图（EEG）和机器学习（ML）模型预测健康成人驾驶员的驾驶状态；并解释与每种驾驶状况相关的神经活动。方法使用Cognionics Quick-20脑电图头戴设备采集被试在驾驶模拟器内360全屏真实车厢内静息状态、城市道路、高速公路和郊区道路驾驶状态下的EEG数据。由于不同的驾驶环境，参与者在驾驶时经历了不同的认知负荷。提取脑电特征，包括谱带功率和功率比，提取与驾驶状态相关的神经活动模式。基于脑电特征，采用梯度增强（GBoost）和随机森林（RF）机器学习分类器对驾驶状态进行分类。为了提高模型的可解释性，采用了一种模型不可知的解释方法，突出了脑电频谱特征对每个驱动状态的影响。结果GBoost模型的分类准确率最高，静息状态的ROC-AUC值为1.00，城郊道路状态为0.96，高速公路状态为0.91，城市道路状态为0.90，能够有效地利用EEG特征识别驾驶状态。模型不可知的特征归因方法突出了关键的EEG贡献因素，如Theta和Delta波段的功率，以及功率比（DAR， DTR），提供了可解释性，并与已建立的认知负荷神经学指标保持一致。本研究证明了基于脑电图的特征与可解释机器学习相结合在驾驶员状态预测中的潜力。该方法为个性化、反应灵敏的驾驶辅助系统奠定了基础，通过监测驾驶员的认知状态来改善道路安全和预防残疾。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Medical Engineering & Physics 工程技术-工程：生物医学

CiteScore

4.30

自引率

4.50%

发文量

172

审稿时长

3.0 months

期刊介绍： Medical Engineering & Physics provides a forum for the publication of the latest developments in biomedical engineering, and reflects the essential multidisciplinary nature of the subject. The journal publishes in-depth critical reviews, scientific papers and technical notes. Our focus encompasses the application of the basic principles of physics and engineering to the development of medical devices and technology, with the ultimate aim of producing improvements in the quality of health care.Topics covered include biomechanics, biomaterials, mechanobiology, rehabilitation engineering, biomedical signal processing and medical device development. Medical Engineering & Physics aims to keep both engineers and clinicians abreast of the latest applications of technology to health care.