Monitoring of sedation in mechanically ventilated patients using remote technology

Area of Science:

• Critical Care Medicine
• Neuroscience
• Medical Technology

Background:

• The COVID-19 pandemic revealed high rates of drug-induced coma and delirium in mechanically ventilated patients.
• High sedative doses, similar to general anesthesia, are often used in acute respiratory distress syndrome without clear electroencephalogram (EEG) endpoints.
• Neurological examination alone cannot detect all pathological brain states, such as burst suppression.

Purpose of the Study:

• To address the challenge of preventing oversedation in the ICU.
• To explore the potential of remote, automated EEG analysis for improved sedation monitoring.
• To enhance the management of mechanically ventilated patients through real-time alerts for abnormal brain activity.

Main Methods:

• Continuous limited-channel EEG monitoring.
• Machine learning-based analysis of EEG signals.
• Development of automated systems for detecting severe EEG suppression and other abnormal brain states.

Main Results:

• Machine learning analysis of continuous EEG signals is feasible.
• Automated EEG analysis can identify burst suppression and delirium with high specificity.
• Remote technologies offer potential for real-time clinical alerts.

Conclusions:

• Continuous EEG monitoring and automated analysis can help prevent oversedation.
• Automated EEG analysis may reduce drug-induced coma and improve patient outcomes in the ICU.
• Remote EEG monitoring technologies can enhance standard care for ventilated patients.