Automated Pupillometry Measurements for Delirium Detection in Surgical ICU Patients

A prospective multicenter observational study collected 417 automated pupillometry measurements from 49 adult surgical intensive care unit patients between April 2023 and August 2025. The data includes pupillometric parameters, delirium assessments using CAM-ICU, and concurrent pain intensity scores. The dataset was created by Patrik Mica and is licensed for open use under CC-BY-4.0.

Use Cases

• Train a classification model to predict delirium based on pupillary light reflex latency.
• Analyze the confounding effect of pain intensity scores on pupillometric parameters.
• Validate automated pupillometry as a diagnostic tool for delirium in a high-pain surgical population.
• Build a mixed-effects model to account for repeated measurements within individual patients.

Strengths

• Includes 417 measurements from 49 patients, providing repeated observations per subject.
• Data collection was prospective and multicenter, likely increasing generalizability.
• Key variables like pupillary latency were independently associated with delirium in a model adjusted for sedation and pain.

Limitations

• The dataset is very small at 16.6 KB, indicating limited scope.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

figshare

Collection Method

Prospective multicenter observational study with repeated measurements.

Time Range

April 2023 to August 2025

Freshness

Last updated 2026-06-04 05:34:49; freshness should be verified.