Design and Testing of an Adjustable Functional Near-Infrared Spectroscopy (fNIRS)

Overview

Functional near-infrared spectroscopy (fNIRS) is a valuable neuroimaging tool due to its safety, portability, and suitability for diverse populations. However, traditional fixed-size fNIRS caps often produce poor data quality, prolonged setup times, and pa rticipant discomfort, particularly for individuals with darker skin tones or thick hair textures. This study introduces a novel adjustable fNIRS cap featuring a 360-degree dial-lacing system, flexible neoprene-nylon fabric, and modular 3D-printed optode ho lders to ensure precise and consistent sensor placement. The hypothesis was that this adjustable cap would offer similar or better data quality compared to standard fixed- size caps due to its tailored fit, while also reducing setup times and enhancing user comfort. To test this hypothesis, the adjustable cap was compared with the commonly used EasyCap in a pilot study involving 15 adult participants with varying head sizes (51–58 cm), hair types, and skin tones (Fitzpatrick scale 1–5). The adjustable cap achieved a significantly higher average optimal (Green) signal quality (82.3%) compared to the EasyCap (45.4%). Improvements were especially evident among participants with darker skin tones (24.2% higher) and curly hair textures (55.8% higher). Additionally, the setup time was dramatically reduced from 25 –30 minutes to seconds, and participants reported increased comfort and reduced motion artifacts. These findings support the hypothesis and highlight the adjustable cap’s potential to improve inclusivity, reduce logistical burdens, lower costs, and enhance data quality in neuroimaging research and clinical practice. Future studies should further validate these benefits across broader populations. Advancing Epileptic Seizure Prediction using a Machine Learning-Based Wearable Device Rebecca Jacob Solon High School, Solon, OH Epilepsy, a brain disorder causing recurring seizures, affects approximately 65 million people worldwide, making it one of the most common neurological diseases. Additionally, epilepsy severely reduces individuals' quality of life, with a disability -adjusted life year (DALY ) of 0.657 – where 1 represents full disability–, indicating a substantial reduction in health. There is a need to predict seizures in real time, and warn patients or caregivers of imminent seizures. I hypothesized that machine learning, implemented in a wea rable device, could achieve this. Using two datasets—CHB-MIT and Siena Scalp EEG —comprising 1160 hours of data from patients aged 0.5 to 71, four models were trained for seizure detection (K -Nearest Neighbors, Logistic Regression, Random Forest Classifier, Support Vector Machine) and one for prediction (Long Short-Term Memory). Evaluation metrics, including accuracy, recall, precision, and F1 -score, showed a detection accuracy of 98.67%. Subsequently, the model was integrated with a wearable EEG headset (Muse) and a web app was developed that successfully predicts seizures 5 minutes ahead of time, with 84.54% accuracy. This work demonstrates the potential of machine learning to enhance seizure prediction and improve patient safety.

Competition history

• JSHS 2025 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.