Portable Brain-Computer Interface for the Intensive Care Unit Patient Communication Using Subject-Dependent SSVEP...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Critical Care Medicine

Background:

• Intensive Care Unit (ICU) patients often face communication barriers, leading to distress and unrecognized needs.
• Existing communication methods in ICUs can be inconsistent and ineffective.
• Brain-Computer Interface (BCI) systems offer a potential solution but face challenges in real-world applications.

Purpose of the Study:

• To develop and evaluate a portable Brain-Computer Interface system for Intensive Care Unit (ICU) communications (BCI4ICU).
• To enhance communication effectiveness and consistency for ICU patients.
• To address limitations of current BCI systems in clinical environments.

Main Methods:

• Designed a portable BCI system (BCI4ICU) using a wearable EEG cap and an Android app for visual stimuli and data processing.
• Proposed a novel subject-specific Gaussian Mixture Model- (GMM-) based training and adaptation algorithm.
• Incorporated subject-specific information into the SSVEP identification model training and adaptation using GMMs.

Main Results:

• The BCI4ICU system achieved an average identification accuracy of 98.7%.
• Subject-specific GMMs demonstrated effectiveness in improving model performance.
• The system successfully generated high-dimensional supervectors for prediction.

Conclusions:

• The developed BCI4ICU system shows significant promise for improving communication in intensive care settings.
• The novel GMM-based training and adaptation algorithm enhances BCI performance in real-world scenarios.
• This technology can lead to more consistent and effective communication for critically ill patients.