A biomedical approach to retrieve information on driver's fatigue by integrating EEG, ECG and blood biomarkers during simulated driving session

A critical analysis of physical fatigue from prolonged driving in real time and subsequent scoring will be a boon for transport sector to prevent road traffic accidents. The current study has two objectives, first, to conduct a multidimensional analysis of central and physical components of fatigue in drivers during simulated driving session and second, to find the rationality of each assessed parameter to be used for scoring system. Briefly, 12 skilled drivers were subjected to simulated driving session for 32-hours. An EEG and an ECG were obtained from each subject at 3-hours interval to assess central and peripheral components of driver's fatigue respectively. EEG data were analyzed to obtain the variation in relative energy of all frequency bands while ECG data were used to study heart rate variability(HRV) at progressive stages of fatigue. Concurrently, blood samples of each subject were analyzed for key blood biomarkers(random blood sugar, blood urea and creatinine) at 8-hours interval. The relative-energy of θ, α and β-bands increased most significantly at Cz electrode while the variations across the stages was most significant in a derived band i.e. α+θ / δ1-δ2 followed by that of θ-band. The power distribution in high frequency components of ECG showed a distinct decreasing trend with advancing fatigue. All blood biomarkers increased with duration of driving task that was significant across the stages. Thus, an effective scoring of drivers' fatigue can be obtained by integrating EEG parameters with HRV and blood biomarkers that can validate fatigue detecting devices under development.