AI-Based Precision Transfusion Prediction Model in Critically Ill Patients

Artificial Intelligence-Based Precision Transfusion Prediction Model for Prevention of Multiple Organ Dysfunction Syndrome in Critically Ill Patients: A Multicenter Observational Study

This multicenter observational study aims to develop and validate an artificial intelligence-based precision transfusion prediction model for critically ill patients. The study will collect clinical characteristics, laboratory parameters, transfusion-related information, physiological data, and clinical outcomes from critically ill patients admitted to intensive care units. An AI model will be developed using retrospective data and further evaluated using prospective observational data. The primary objective is to investigate factors associated with multiple organ dysfunction syndrome (MODS) and establish a predictive model to support individualized transfusion management in critically ill patients.

Study Overview

• Status: Not yet recruiting
• Conditions: Critical Illness; Blood Transfusion; Multiple Organ Dysfunction Syndrome
• Intervention / Treatment: Other: Red Blood Cell Transfusion Exposure

Detailed Description

Critically ill patients frequently require red blood cell transfusion during intensive care. However, transfusion decisions based solely on conventional indicators may not fully reflect individual differences in disease severity, oxygen delivery, and risk of organ dysfunction. Unnecessary transfusion may increase the risk of adverse outcomes, whereas delayed transfusion may worsen tissue hypoxia.

This multicenter observational study aims to establish an artificial intelligence-based precision transfusion prediction model for critically ill patients. The study includes retrospective model development and prospective observational validation phases.

Clinical data including demographic characteristics, underlying diseases, laboratory parameters, physiological variables, transfusion records, severity scores, organ function indicators, and clinical outcomes will be collected. Machine learning approaches will be applied to identify important predictors associated with multiple organ dysfunction syndrome (MODS) and transfusion-related outcomes.

The developed model will be evaluated based on predictive performance, including discrimination, calibration, and clinical applicability. This study aims to provide an individualized risk assessment approach to improve transfusion decision-making and facilitate precision management in critically ill patients.

• Study Type: Observational
• Enrollment (Estimated): 2598

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: N/A

• Sampling Method: Non-Probability Sample

Study Population

This study includes adult critically ill patients admitted to intensive care units. Clinical data including demographic characteristics, underlying diseases, severity indicators, laboratory parameters, transfusion-related information, and clinical outcomes will be collected from electronic medical records. The study population will be used to develop an artificial intelligence-based prediction model for precision transfusion management and risk assessment of multiple organ dysfunction syndrome in critically ill patients.

Description

Inclusion Criteria:

• Adult patients (aged ≥18 years) admitted to the intensive care unit.
• Patients with available clinical data, including demographic characteristics, laboratory parameters, transfusion-related information, and clinical outcomes.
• Patients meeting the requirements for model development and analysis.

Exclusion Criteria:

• Patients younger than 18 years.
• Patients with missing key clinical information required for analysis.
• Patients with repeated ICU admissions during the study period (only the first ICU admission will be included).
• Patients whose data cannot be used for research purposes according to ethical requirements.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Critically Ill Patient Cohort

Other: Red Blood Cell Transfusion Exposure; Red blood cell transfusion exposure refers to the receipt of red blood cell transfusion during intensive care hospitalization. Transfusion-related information, including transfusion status, number of transfused units, and cumulative transfusion volume, will be collected from routine clinical care records. Transfusion decisions are not assigned by the study protocol, and no intervention is performed as part of this observational study.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Occurrence of Multiple Organ Dysfunction Syndrome (MODS)	The primary outcome is the occurrence of multiple organ dysfunction syndrome (MODS) during intensive care unit hospitalization. MODS will be defined according to clinical diagnostic criteria based on dysfunction of two or more organ systems.	During ICU hospitalization (up to 28 days after ICU admission)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
ICU Mortality	Death occurring during the intensive care unit stay.	Up to Day 28 after ICU admission
Hospital Mortality	Death occurring during the index hospitalization	Up to Day 28 after ICU admission

Collaborators and Investigators

• Sponsor: Second Affiliated Hospital, Zhejiang University, School of Medicine

Study record dates

Study Major Dates

• Study Start (Estimated): September 1, 2026
• Primary Completion (Estimated): May 1, 2029
• Study Completion (Estimated): September 1, 2029

Study Registration Dates

• First Submitted: August 5, 2026
• First Submitted That Met QC Criteria: August 11, 2026
• First Posted (Actual): August 13, 2026

Study Record Updates

• Last Update Posted (Actual): August 13, 2026
• Last Update Submitted That Met QC Criteria: August 11, 2026
• Last Verified: August 1, 2026

More Information

Terms related to this study

• Keywords: Artificial Intelligence; Critically Ill Patients; Multiple Organ Dysfunction Syndrome; Precision Transfusion
• Additional Relevant MeSH Terms: Pathologic Processes; Disease Attributes; Shock; Pathological Conditions, Signs and Symptoms; Critical Illness; Multiple Organ Failure
• Other Study ID Numbers: 20260140; IR2025639 (Other Identifier: Human Research Ethics Committee of The Second Affiliated Hospital, Zhejiang University School of Medicine)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO