Artificial intelligence and machine learning in disorders of consciousness

Area of Science:

• Neuroscience
• Medical Informatics

Background:

• Severe disorders of consciousness (DoC) following acquired brain damage present diagnostic and prognostic challenges.
• Artificial intelligence (AI) and machine learning (ML) offer promising tools to improve patient care.

Purpose of the Study:

• To review recent studies utilizing AI and ML to reduce diagnostic and prognostic uncertainty in DoC.
• To characterize patient responses to novel therapeutic interventions using AI/ML.

Main Methods:

• Analysis of functional neuroimaging and electroencephalography (EEG) data using AI/ML.
• Application of conventional machine learning and deep learning algorithms.
• Utilizing Glasgow Outcome Scale for outcome prediction.

Main Results:

• AI/ML effectively differentiate between unresponsive wakefulness syndrome and minimally conscious state.
• Machine learning models predict patient outcomes and responses to interventions like zolpidem and tDCS.
• Most studies employed conventional ML over deep learning for analysis.

Conclusions:

• AI and ML are valuable tools for clinical decision-making in DoC diagnosis, prognosis, and therapy.
• Deep learning techniques hold significant potential to enhance the performance of AI/ML models in DoC management.