Machine Learning-based Longitudinal Study of Post-ICU Syndrome Development Trajectory in Critically Ill Patients and Construction of Clinical Early Warning Models: a Research Protocol for Longitudinal Study

Overview

This project intends to track and evaluate whether post-ICU syndrome will occur 7 days, 1 month, 3 months and 6 months after ICU patients are transferred out of the ICU through a longitudinal study, apply the latent category growth model to identify different trajectory patterns of post-ICU syndrome in critically ill patients, and use modern machine learning models to build an early warning model of the trajectory patterns of post-ICU syndrome.

Eligibility

Inclusion Criteria:

• Length of stay in ICU ≥24h; ②Age ≥18 years old; ③ Conscious when leaving ICU, communicating with investigators without barriers; ④ Informed consent.

Exclusion Criteria:

• ① had been admitted to ICU for more than 24h within 3 months prior to this admission; ② Transferred to another ICU; ③ There was cognitive impairment before ICU admission (BDRS > 4); (4) Serious hearing impairment, dysarthria, etc., can not be followed up; ⑤ Serious illness can not cooperate to complete the questionnaire.