Early Delirium Prediction Via Serial EEG Trajectories and Machine Learning

NCT07536854 | Completed

• 1 other identifier: AJOUIRB-IV-2024-227
• Study Type: observational
• Target: 73
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this observational study is to develop a machine learning model that can predict delirium in trauma patients before it clinically appears. The study focuses on analyzing brainwave (EEG) patterns collected over several days in the trauma ICU. By comparing different recording conditions-such as having eyes open versus closed-researchers aim to identify the most effective way to monitor brain health and detect early signs of delirium in critically ill patients.

Conditions

Delirium; Trauma; Brain Dysfunction; Critical Illness

Keywords

delirium; EEG; machine learning; Delirium prediction

Outcome Measures

Primary Outcomes (1)

• Predictive Performance for Delirium (Area Under the Receiver Operating Characteristic Curve, AUROC

  The predictive accuracy of the machine learning model based on longitudinal EEG trajectories will be evaluated to identify patients at risk of delirium. Model performance will be assessed using AUROC, sensitivity, specificity, and F1-score.

  3 to 4 days (during the longitudinal EEG data collection period)

Secondary Outcomes (1)

• Comparison of Model Performance: Eyes-Open vs. Eyes-Closed States

  3 to 4 days

Study Arms (2)

Delirum group

Patients who developed delirium during their ICU stay (confirmed by CAM-ICU)

Non-Delirium Group

Patients who did not develop delirium during their ICU stay.

Eligibility Criteria

• Age: 18 Years - 65 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult trauma patients (aged 18-65) admitted to the Trauma Intensive Care Unit (TICU) at a level 1 trauma center in Korea. The cohort includes critically ill patients with a record of severe injury (ISS≥ 9) who are able to undergo serial EEG monitoring and clinical delirium assessments. Patients with pre-existing neurological or psychiatric disorders, or those with severe traumatic brain injury (AIS ≥ 2), are excluded to ensure the specificity of the neurophysiologic data.

You may qualify if:

• Trauma patients admitted to the Trauma Intensive Care Unit (TICU) who meet the following criteria:
• Patients aged 18 to 65 years.
• Severe trauma patients with an Injury Severity Score (ISS)

You may not qualify if:

• Patients with a head Abbreviated Injury Scale (AIS) ≥ 2 Patients with a Richmond Agitation-Sedation Scale (RASS) score ≤ -2 History of neurological disorders (e.g., Parkinson's disease, dementia, cerebrovascular disease) History of major psychiatric disorders (e.g., schizophrenia, bipolar disorder, intellectual disability, autism spectrum disorder) History of illicit drug use disorder or positive results on a urine drug screen for substances other than Benzodiazepines or Tricyclic antidepressants.
• Clinical evidence of acute alcohol withdrawal (CIWA-Ar score \> 10) History of liver failure or hepatic encephalopathy (Child-Pugh Class B or C) Renal impairment requiring renal replacement therapy (RRT) Inability to perform the Confusion Assessment Method for the ICU (CAM-ICU) due to the following Inability to communicate in Korean Failure to obey commands (unable to follow test instructions) Severe visual or hearing impairment Refusal to undergo CAM-ICU assessment Requirement for isolation due to infectious diseases (e.g., COVID-19, active tuberculosis).

Sponsors & Collaborators

• Ajou University School of Medicine: lead

Study Sites (1)

Ajou University Hospital

Suwon, Kyonggi-do, 16499, South Korea

Location

Related Publications (6)

• Sun H, Kimchi E, Akeju O, Nagaraj SB, McClain LM, Zhou DW, Boyle E, Zheng WL, Ge W, Westover MB. Automated tracking of level of consciousness and delirium in critical illness using deep learning. NPJ Digit Med. 2019 Sep 9;2:89. doi: 10.1038/s41746-019-0167-0. eCollection 2019.

  PMID: 31508499 BACKGROUND

• Numan T, van den Boogaard M, Kamper AM, Rood PJT, Peelen LM, Slooter AJC; Dutch Delirium Detection Study Group. Delirium detection using relative delta power based on 1-minute single-channel EEG: a multicentre study. Br J Anaesth. 2019 Jan;122(1):60-68. doi: 10.1016/j.bja.2018.08.021. Epub 2018 Oct 2.

  PMID: 30579407 BACKGROUND

• Kim H, McKinney A, Brooks J, Mashour GA, Lee U, Vlisides PE. Delirium, Caffeine, and Perioperative Cortical Dynamics. Front Hum Neurosci. 2021 Dec 20;15:744054. doi: 10.3389/fnhum.2021.744054. eCollection 2021.

  PMID: 34987367 BACKGROUND

• Hshieh TT, Saczynski J, Gou RY, Marcantonio E, Jones RN, Schmitt E, Cooper Z, Ayres D, Wright J, Travison TG, Inouye SK; SAGES Study Group. Trajectory of Functional Recovery After Postoperative Delirium in Elective Surgery. Ann Surg. 2017 Apr;265(4):647-653. doi: 10.1097/SLA.0000000000001952.

  PMID: 27501176 BACKGROUND

• Bryczkowski SB, Lopreiato MC, Yonclas PP, Sacca JJ, Mosenthal AC. Risk factors for delirium in older trauma patients admitted to the surgical intensive care unit. J Trauma Acute Care Surg. 2014 Dec;77(6):944-51. doi: 10.1097/TA.0000000000000427.

  PMID: 25248058 BACKGROUND

• Walder B, Haase U, Rundshagen I. [Sleep disturbances in critically ill patients]. Anaesthesist. 2007 Jan;56(1):7-17. doi: 10.1007/s00101-006-1086-4. German.

  PMID: 16953422 BACKGROUND

MeSH Terms

Conditions

Delirium; Wounds and Injuries; Critical Illness

Condition Hierarchy (Ancestors)

Confusion; Neurobehavioral Manifestations; Neurologic Manifestations; Nervous System Diseases; Signs and Symptoms; Pathological Conditions, Signs and Symptoms; Neurocognitive Disorders; Mental Disorders; Disease Attributes; Pathologic Processes

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: PROSPECTIVE
• Target Duration: 3 Days
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Clinical associate professor

Study Record Dates

• First Submitted: April 12, 2026
• First Posted: April 17, 2026
• Study Start: April 1, 2024
• Primary Completion: April 27, 2025
• Study Completion: April 30, 2025
• Last Updated: April 17, 2026

Record last verified: 2026-04

Data Sharing

• IPD Sharing: Will not share

Locations

KR (1)