Illustration of a tired driver with EEG brainwaves in the background, symbolizing fatigue detection.

Driving Drowsy? How New Tech Can Spot Fatigue Before It's Too Late

Author's portrait.
Theo Raines in Health & Wellness June 2025 • 5 min read.

"An automated EEG analysis algorithm offers a promising solution for detecting and mitigating driver fatigue, potentially preventing accidents related to obstructive sleep apnea and other sleep disorders."


Driving can be a risky task, and feeling sleepy makes it much worse. Driver fatigue is a big problem, causing up to 20% of serious car accidents. Conditions like obstructive sleep apnea (OSA), where people stop breathing during sleep, make this even more dangerous by causing severe fatigue and drowsiness. People with OSA are up to five times more likely to crash. Figuring out who is most at risk and predicting accidents is a big challenge.

Researchers are looking for ways to monitor how alert drivers are. Analyzing brain waves (EEG) can show how sleepy someone is, which affects their driving. When people are tired, their EEG shows more activity in certain frequency bands like alpha and theta. This method helps keep an eye on drivers' awareness and how well they're doing behind the wheel. However, getting reliable EEG data while someone is driving is tough because there can be a lot of interference.

One of the biggest problems in using EEG for this purpose is dealing with artifacts—things like movements, blinks, and electrical noise that mess up the brainwave signals. A technique called independent component analysis (ICA) helps to clean up these signals, but often it's not enough. Additional manual cleaning is needed, which takes a lot of time and can be subjective. To solve this, a team of researchers developed a new automated method to detect and remove these artifacts more efficiently.

How Does the Automated Algorithm Work?

Illustration of a tired driver with EEG brainwaves in the background, symbolizing fatigue detection.

The study, published in the International Journal of Psychophysiology, tested a new automated algorithm (AA) designed to detect and remove artifacts from EEG recordings taken during a driving simulation. The goal was to see if this method could accurately identify and eliminate noise, making it easier to analyze brainwave data and assess driver fatigue. Researchers compared the automated algorithm against a reference-standard (RS) method, where experts visually inspected the EEG data for artifacts.

The study involved five patients with obstructive sleep apnea (OSA) and five healthy controls. Participants underwent a 40-hour extended wakefulness study, which included 30-minute simulated driving tasks every two hours. During these tasks, EEG data was continuously recorded. The EEG data was processed in the following stages:

  • Initial Processing with ICA: The raw EEG data was first cleaned using independent component analysis (ICA) to remove basic artifacts like eye blinks.
  • Automated Artifact Detection: The automated algorithm then analyzed the ICA-processed data to detect any remaining artifacts. This was done by looking at the standard deviation of EEG amplitude (SDEA) in each channel within 5-second epochs. If the SDEA exceeded a set threshold, the epoch was flagged as artefactual.
  • Reference-Standard (Visual Inspection): A sleep expert visually inspected the EEG data to identify epochs containing artifacts, serving as the reference for evaluating the automated algorithm.
  • Comparison and Analysis: The performance of the automated algorithm was assessed by comparing its results with the reference-standard. Sensitivity, specificity, and accuracy were calculated to determine how well the algorithm performed.
After identifying and removing noisy epochs, EEG spectral power was calculated using three methods: the reference-standard (RS), the automated algorithm (AA), and raw data with only ICA for artifact rejection. This allowed the researchers to compare the effectiveness of each method in providing clean, usable EEG data.

The Future of Driver Safety Technology

This research offers a promising step forward in using technology to improve driver safety. By automating the process of EEG artifact removal, the study makes it easier to gather and analyze brainwave data in real-world driving scenarios. This could lead to new ways of detecting and preventing driver fatigue, especially in people with conditions like obstructive sleep apnea. As technology advances, we may see more tools like this integrated into vehicles, helping to keep drivers alert and roads safer for everyone.

About this Article -

This article was crafted using a human-AI hybrid and collaborative approach. AI assisted our team with initial drafting, research insights, identifying key questions, and image generation. Our human editors guided topic selection, defined the angle, structured the content, ensured factual accuracy and relevance, refined the tone, and conducted thorough editing to deliver helpful, high-quality information.See our About page for more information.

Everything You Need To Know

1

How significant is driver fatigue as a contributing factor to car accidents, and what role does obstructive sleep apnea play in increasing this risk?

Driver fatigue contributes to approximately 20% of serious car accidents. Conditions such as obstructive sleep apnea (OSA) exacerbate this risk, increasing the likelihood of accidents by up to five times. Predicting and identifying at-risk drivers remains a significant challenge, driving the need for innovative monitoring solutions.

2

Can you explain the core mechanism of the new automated algorithm for detecting artifacts in EEG data during driving simulation?

The automated algorithm uses the standard deviation of EEG amplitude (SDEA) to automatically detect and remove artifacts from EEG recordings during driving simulations. It assesses the SDEA in each channel within 5-second epochs. If the SDEA exceeds a predefined threshold, the epoch is flagged as containing artifacts and is removed. This is done after an initial cleaning using independent component analysis (ICA) to remove basic artifacts like eye blinks.

3

How did the study validate the effectiveness of the new automated algorithm against existing methods for EEG artifact removal?

The study compared the new automated algorithm (AA) with a reference-standard (RS) method. In the reference standard, a sleep expert visually inspected EEG data to identify epochs containing artifacts. The performance of the automated algorithm was then evaluated by comparing its results with this reference standard. Sensitivity, specificity, and accuracy were calculated to determine the algorithm's effectiveness in detecting and removing artifacts.

4

What were the different methods used to compute EEG spectral power and why was it important to compare them in this study?

The researchers computed EEG spectral power using three distinct methods: the reference-standard (RS), the automated algorithm (AA), and raw data processed only with independent component analysis (ICA) for artifact rejection. This comparative analysis allowed the assessment of each method's effectiveness in yielding clean and usable EEG data, essential for accurate analysis of driver fatigue.

5

What potential future impacts might the integration of technologies, such as the automated algorithm, have on driver safety and accident prevention?

Future integration of technologies like the automated algorithm into vehicles could revolutionize driver safety by proactively detecting and mitigating driver fatigue, especially in individuals with conditions like obstructive sleep apnea (OSA). Such advancements may lead to a significant reduction in accidents related to fatigue and drowsiness, paving the way for safer roads and a decrease in traffic-related fatalities.

Newsletter Subscribe

Subscribe to get the latest articles and insights directly in your inbox.