Disorders of consciousness (DoC) are severe neurological conditions characterized by impaired awareness and reduced interaction with the environment. Although hyperbaric oxygen therapy (HBOT) may promote neurological recovery, the optimal treatment strategy for individual patients remains unclear.
This study aims to develop and validate an artificial intelligence-assisted precision decision support system for HBOT based on dynamic BIS-EEG brain functional feedback. The system will integrate clinical information, neuroimaging data, physiological parameters, consciousness assessments, and HBOT treatment characteristics.
Continuous BIS-EEG monitoring will be performed during different phases of HBOT, including compression, oxygen exposure, air breaks, decompression, and post-treatment recovery. Artificial intelligence models will analyze these dynamic brain functional responses to predict treatment effectiveness and support personalized HBOT management.
The primary outcome will be improvement in consciousness assessed by the Coma Recovery Scale-Revised (CRS-R). Secondary outcomes include changes in consciousness state, cognitive and functional recovery, and treatment-related adverse events.
Adults aged 18 years or older.
Diagnosis of disorders of consciousness, including vegetative state/unresponsive wakefulness syndrome (VS/UWS) or minimally conscious state (MCS).
Disease duration of at least 28 days after the onset of brain injury.
Stable medical condition suitable for hyperbaric oxygen therapy.
Contraindications to hyperbaric oxygen therapy, including untreated pneumothorax or other conditions considered unsafe for HBOT.
Severe cardiovascular, respiratory, or systemic instability.
Uncontrolled epilepsy or frequent seizures affecting assessment.
Severe infection or metabolic disorders that may influence consciousness assessment.
Parallel
AI-Assisted Personalized Hyperbaric Oxygen Therapy
Other interventionParticipants will receive hyperbaric oxygen therapy optimized by an artificial intelligence-assisted clinical decision support system. The system integrates dynamic BIS-EEG brain functional feedback, clinical characteristics, neuroimaging findings, physiological parameters, and previous treatment responses to provide individualized recommendations for hyperbaric oxygen therapy parameters, including treatment pressure, oxygen exposure duration, air-break strategy, treatment frequency, and treatment course. Final treatment decisions will be made by qualified physicians.
Conventional Hyperbaric Oxygen Therapy
Other interventionParticipants will receive conventional hyperbaric oxygen therapy according to standard clinical protocols determined by treating physicians. Treatment parameters will be selected based on routine clinical practice without artificial intelligence-assisted recommendations.
Change From Baseline in Coma Recovery Scale-Revised (CRS-R) Total Score at Week 12
The Coma Recovery Scale-Revised (CRS-R) will be used to assess the level of consciousness. The CRS-R total score ranges from 0 to 23, with higher scores indicating better neurobehavioral function and a higher level of consciousness. The change in CRS-R total score from baseline to Week 12 will be calculated for each participant and compared between the AI-assisted precision hyperbaric oxygen therapy group and the conventional hyperbaric oxygen therapy group.
Change From Baseline in Glasgow Coma Scale (GCS) Total Score at Week 12
The Glasgow Coma Scale (GCS) will be used to assess neurological responsiveness. The total score ranges from 3 to 15, with higher scores indicating better neurological responsiveness. Change from baseline to Week 12 will be compared between the AI-assisted precision hyperbaric oxygen therapy group and the conventional hyperbaric oxygen therapy group.
Number of Participants With Hyperbaric Oxygen Therapy-Related Adverse Events
The number and proportion of participants experiencing adverse events considered related to hyperbaric oxygen therapy will be recorded and compared between groups. Events of interest include middle-ear or sinus barotrauma, oxygen toxicity, seizures, pulmonary symptoms, hemodynamic instability, and other clinically significant treatment-related adverse events.
Discrimination Performance of the AI Model for Predicting Improvement in Consciousness State
The predictive performance of the artificial intelligence model will be evaluated for predicting improvement to a higher consciousness state at Week 12. Discrimination will be assessed using the area under the receiver operating characteristic curve (AUROC). An AUROC of 0.5 indicates no discrimination, whereas values closer to 1.0 indicate better discrimination.
Click on the lead sponsor to view all of their trials.
This trial is not recruiting at the moment. You can still explore other options:
