Wearable fNIRS Sensor Tracks Cognitive Fatigue in Real Time

Wireless fNIRS sensor concept wearable headband that monitors brain activity in real time © stefanholm-chronicles-stock.adobe.com

Wearable Sensor Brings Brain Monitoring Out of the Lab

In a leap forward for portable neuroscience and real-time cognitive tracking, a team of engineers has unveiled a wearable wireless system that uses near-infrared spectroscopy (NIRS) to monitor brain activity linked to cognitive fatigue. The lightweight, Bluetooth-enabled sensor—about the size of a bandage—can detect changes in brain oxygenation and transmit data to a smartphone or laptop, opening the door for AI-driven diagnostics in real-world settings (1,2).

Developed by researchers Mauro Victorio, James Dieffenderfer, Tanner Songkakul, Josh Willeke, Alper Bozkurt, and Vladimir A. Pozdin from Florida International University, North Carolina State University, and Rose-Hulman Institute of Technology, the new system focuses on the prefrontal cortex (PFC)—a region key to decision-making, memory, and higher-order thinking (1). Their findings, now posted as a preprint on Preprints.org (1), demonstrate that the compact device delivers reliable hemodynamic data, aligning closely with traditional benchtop NIRS and even fMRI benchmarks (1).

Optics and Engineering for the Brain

Traditional brain imaging techniques like functional magnetic resonance imaging (fMRI) provide high-resolution maps of brain function but are expensive, non-portable, and impractical for continuous monitoring. Functional near-infrared spectroscopy (fNIRS), by contrast, tracks changes in oxy- and deoxy-hemoglobin in the brain using light in the 650–850 nm range. It is non-invasive, relatively low-cost, and—crucially—can be miniaturized (1,2).

The new sensor system features a two-layer printed circuit board with skin-facing LEDs and photodetectors, while all other electronics are placed on the reverse. A 3D-printed flexible shroud enhances comfort and blocks ambient light, and the full assembly is coated in biocompatible Parylene-C. Key to the device’s performance is an optical coupling strategy using polydimethylsiloxane (PDMS), which significantly improves light transmission into the skin, especially at larger optode separations (1).

Real-Time Hemodynamic Feedback and Cognitive Testing

In validation tests, the device successfully tracked physiological responses to an arterial occlusion test and breath-holding exercise. As blood flow was restricted, the sensor recorded expected drops in oxygenated hemoglobin and spikes in deoxygenated hemoglobin, followed by recovery upon release—demonstrating fidelity comparable to full-scale lab equipment.

When placed on the prefrontal cortex, the sensor captured real-time brain oxygenation changes during arithmetic tasks. Some participants showed increased oxygenation with cognitive effort, while others displayed a decrease, particularly under higher workloads. Notably, one participant’s signal shifted mid-study—an indicator, the authors suggest, of progressing cognitive fatigue. These distinctions were detectable without post-processing, highlighting the device’s potential for edge-computing applications (1).

AI and Frequency Domain Insights

Though the system currently operates with basic signal processing, it is designed for future integration with artificial intelligence (AI) and machine learning (ML). High-frequency and very low frequency (VLF) components of the hemodynamic signal—down to 2.2 mHz—were examined to distinguish mental workload from noise. Participants experiencing fatigue exhibited persistent VLF shifts that extended into rest periods, consistent with known fatigue biomarkers (1).

The authors noted that yawning, often dismissed in such studies, may correlate with transient spikes in signal frequency and even influence brain oxygenation, though its effects remain inconclusive.

Toward Everyday Cognitive Monitoring

Despite some limitations—including small participant groups and variations in verbal task completion—the study makes a compelling case for wearable fNIRS as a viable method for cognitive workload monitoring in mobile settings. The PFC’s accessibility and physiological activity levels make it a prime target for such applications, whether in autonomous vehicles, aerospace cockpits, or classrooms (1).

Ultimately, this research lays the foundation for wearable brain analytics augmented by AI. With scalable signal fidelity and data-rich output, the device offers a promising platform for real-time mental state assessment across diverse environments.

References:

(1) Victorio, M.; Dieffenderfer, J.; Songkakul, T.; Willeke, J.; Bozkurt, A.; Pozdin, V. A. Wearable Wireless Functional Near Infrared Spectroscopy System for Cognitive Activity Monitoring. Biosensors 2025, 15 (2), 92. DOI: 10.20944/preprints202501.1283.v1

(2) Gong, Y.; Lin, R.; Mutlu, M. C.; Lorentz, L.; Shaikh, U. J.; Graeve, J. D.; Badkoubeh, N.; Zeng, R. R.; Klein, F.; Lührs, M.; Mathiak, K. The Use of Functional Near-Infrared Spectroscopy (fNIRS) for Monitoring Brain Function, Predicting Outcomes, and Evaluating Rehabilitative Interventional Responses in Poststroke Patients with Upper Limb Hemiplegia: A Systematic Review. IEEE J. Sel. Top. Quantum Electron. 2025, n/a, Article in Press. DOI: 10.1109/JSTQE.2025.3563153