NCT02465307

Intelligent Intensive Care Unit

Motion Analysis of Delirium in Intensive Care Units (ICUs) Subtitle 1: "ADAPT: Autonomous Delirium Monitoring and Adaptive Prevention"

• Status: Active Not Recruiting
• Phase: —
• Study type: Observational
• Enrollment: 130 (estimated)
• Sponsor: University of Florida · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Accepted

Summary

Delirium, as a common complication of hospitalization, poses significant health problems in hospitalized patients. Though about a third of delirium cases can benefit from intervention, detecting and predicting delirium is still very limited in practice. A common characterization of delirium is change in activity level, causing patients to become hyperactive or hypoactive which is manifested in facial expressions and total body movements. This pilot study is designed to test the feasibility of a delirium detection system using movement data obtained from 3-axis wearable accelerometers and commercially available camera with facial recognition video system in conjunction with electronics medical record (EMR) data to analyze the relation of whole-body movement and facial expressions with delirium.

Detailed description

The aim of the study is to assess the potential of using motion and facial expression data to detect delirium in ICU patients by comparing motion and facial expression patterns in delirium and control groups. In this study, the investigators will use ActiGraph accelerometers to record each subject's movement patterns. Also, a processed video using a commercially available camera interfaces with a specialized program to identify patient facial expressions and movement patterns. A total of 60 participants will be enrolled with delirium, and 30 patients without delirium will be used as control group. Motion profiles will be compared in the motorically defined subgroups (hyperactive, hypoactive, normal) based on accelerometer and facial recognition data. Then, differences in facial expression, number of changes in postures, and percentage of time spent moving will be compared between motorically defined subgroups and in delirium and control groups. EMR data will also be used to assess the feasibility of detecting delirium by including additional information on related risk factors.

Conditions

• Delirium
• Confusion

Interventions

Type	Name	Description
BEHAVIORAL	Confusion Assessment Method	Confusion Assessment Method (CAM) score
DEVICE	Accelerometer	3 accelerometers (placed on upper arm, wrist and ankle) and 1 placed on wall as ambient light sensor
DEVICE	Commercially available camera	As part of facial recognition video system
DEVICE	Internet Pod (iPod)	Monitors noise levels in the room
DIAGNOSTIC_TEST	Cortisol Swab	Cortisol level collected through self administered salivary swab

Timeline

• Start date: 2016-02-01
• Primary completion: 2020-03-11
• Completion: 2028-05-30
• First posted: 2015-06-08
• Last updated: 2025-06-29

Locations

1 site across 1 country: United States