Skip to content

A Study on the Application of EEG-Based Multimodal Artificial Intelligence Models in the Standardized Assessment of CRS-R Scores in Patients with Impaired Consciousness

A Study on the Application of EEG-Based Multimodal Artificial Intelligence Models in the Standardized Assessment of CRS-R Scores in Patients with Impaired Consciousness

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600124984
Enrollment
Unknown
Registered
2026-05-20
Start date
2026-06-01
Completion date
Unknown
Last updated
2026-05-25

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Prolonged disorders of consciousness

Interventions

The second phase:None

Sponsors

Guangdong Provincial Second Hospital of Traditional Chinese Medicine (The Fifth Clinical Medical College of Guangzhou University of Chinese Medicine)
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1. Age 18–80 years; 2. Patients with impaired consciousness following severe acquired brain injury, including traumatic brain injury, stroke, hypoxic-ischemic brain injury, postoperative brain injury, encephalitis, or other non-degenerative structural brain injuries; 3. Vital signs are relatively stable, and the patient is capable of completing two independent CRS-R assessments and routine EEG recordings; 4. A legal representative is available to sign a written informed consent form; 5. Expected to be able to complete the study protocol by the time of the baseline assessment.

Exclusion criteria

Exclusion criteria: 1. Persistent deep sedation, general anesthesia, or uncontrolled status epilepticus during the scheduled assessment window; 2. Hemodynamic or respiratory instability that precludes the safe completion of a bedside assessment or EEG examination; 3. Concurrent severe failure of a major organ, or a life expectancy of less than 2 weeks as determined by the investigator; 4. Severe uncorrected bilateral visual or hearing impairment that prevents the administration of standardized behavioral assessments; 5. History of confirmed neurodegenerative disease, severe developmental disorders, or other conditions that significantly affect the interpretation of the CRS-R; 6. Other circumstances deemed inappropriate for enrollment by the investigator.

Design outcomes

Primary

MeasureTime frame
Consistency between AI-predicted CRS-R total scores and the expert-adjudicated gold standard reference;

Secondary

MeasureTime frame
Mean Absolute Error (MAE) of AI-predicted CRS-R total scores;Diagnostic accuracy of the AI model for categorical state of consciousness (UWS/MCS/EMCS);

Countries

China

Contacts

Public ContactChi Xiangfeng

Guangdong Provincial Second Hospital of Traditional Chinese Medicine (The Fifth Clinical Medical College of Guangzhou University of Chinese Medicine)

110988497@qq.com+86 20 8348 2172

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 5, 2026