Eye Tracking for Early Identification of Post-ICU Cognitive Impairment

Post-intensive care syndrome-cognitive impairment (PICS-CI) is a common problem among patients after discharge from the intensive care unit (ICU). It may involve difficulties with memory, attention, executive function, information processing speed, visuospatial ability, or language. These problems may affect long-term recovery, quality of life, treatment adherence, and social participation. Early identification of possible cognitive impairment after ICU discharge is therefore important for follow-up care and rehabilitation planning.

Current assessment of post-ICU cognitive impairment mainly relies on cognitive screening scales and clinical judgment. Although these methods are widely used, they may be influenced by a patient's level of cooperation, education, emotional state, fatigue, and subjective responses. Eye-tracking technology provides objective and quantitative information about eye movement behavior during visual and cognitive tasks. Because eye movements are closely related to attention, inhibitory control, visual processing, and other cognitive processes, eye-tracking-derived parameters may provide additional information to support early cognitive screening after ICU discharge.

This is a prospective, single-center, observational study conducted at Zhongshan Hospital, Fudan University. The study will enroll patients who have been transferred from the ICU to a general ward after ICU treatment. No treatment, medication, device intervention, or change in usual clinical care will be assigned by the study. All study procedures will be performed on the basis of routine care and are intended for research assessment only.

Within 7 days after ICU discharge, participants will complete a standardized cognitive assessment and an eye-tracking test. Cognitive status will be assessed using the Montreal Cognitive Assessment (MoCA), with education-based score correction, together with clinical judgment under the DSM-5 framework for neurocognitive disorders. Based on this assessment, participants will be classified as having post-ICU cognitive impairment or not having post-ICU cognitive impairment.

During the eye-tracking test, participants will complete predefined visual cognitive tasks in a standardized testing environment. Eye movement data will be recorded using eye-tracking equipment and processed using Tobii Pro Lab software according to predefined areas of interest, time windows, and saccade identification rules. Eye-tracking-derived parameters may include saccade latency, saccade accuracy, saccade error rate, mean saccade amplitude, mean saccade velocity, mean saccade duration, fixation time, and uncorrected error rate. These parameters will be used to describe eye movement behavior related to response initiation, attention orientation, inhibitory control, spatial localization, eye movement execution, and visual processing.

The study team will also collect demographic and clinical information from medical records, including age, sex, education level, ICU-related clinical characteristics, disease severity, ICU length of stay, mechanical ventilation, sedative and analgesic exposure, and other clinically relevant factors. These data will be used to describe the study population and to adjust for potential confounding factors in the analysis.

The main objective of this study is to evaluate whether eye-tracking-derived parameters, alone or in combination, can support the early identification of post-ICU cognitive impairment as an auxiliary tool to standard cognitive assessment. The study will compare eye movement characteristics between patients with and without post-ICU cognitive impairment and examine the association between eye-tracking-derived parameters and cognitive status. The diagnostic performance of selected eye-tracking parameters or parameter combinations will be evaluated using receiver operating characteristic curve analysis. Exploratory models may also assess whether adding eye-tracking-derived parameters to clinical and cognitive assessment information improves identification performance.

Eye tracking is non-invasive and does not involve radiation, medication, or invasive procedures. Some participants may experience temporary eye fatigue, visual discomfort, dizziness, nausea, or other mild discomfort while looking at the screen or completing the visual tasks. The research team will monitor participants during testing and may pause or stop the test if needed. All collected data will be de-identified and managed according to the study data management plan to protect participant privacy and data security.