Machine Learning for Transfusion Prediction in Oncologic Surgery

NCT07875114 | Active Not Recruiting DMC

• 1 other identifier: KSH_CA_1
• Study Type: observational
• Target: 950
• Locations: 1 country
• Sites: 1

Brief Summary

This study aims to use artificial intelligence to predict intraoperative bleeding and blood loss in advance for oncologic surgeries with high bleeding risk. The goal is to more accurately determine the type and amount of blood products reserved for patients, improving the efficient use of blood bank resources.

Conditions

Blood Transfusion

Keywords

Machine learning; Perioperative blood transfusion; Oncologic surgery; Blood product prediction; Patient blood management

Outcome Measures

Primary Outcomes (1)

• Prediction of Packed Red Blood Cell (PRBC) Transfusion Requirement Using Artificial Intelligence

  In patients undergoing oncologic surgery, the classification performance of an artificial intelligence algorithm-developed using preoperative clinical and laboratory data-in predicting the need for packed red blood cell (PRBC) transfusion (yes/no) will be evaluated against the patient's actual receipt of PRBC transfusion during the perioperative period (from the start of surgery to 24 hours postoperatively). The area under the receiver operating characteristic curve (AUC-ROC) will serve as the primary metric, with sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) as supportive metrics.

  Perioperative (from the start of surgery to 24 hours postoperatively)

Secondary Outcomes (3)

• Prediction of Fresh Frozen Plasma (FFP) Transfusion Requirement Using Artificial Intelligence

  Perioperative (from the start of surgery to 24 hours postoperatively)

• Comparison of Artificial Intelligence Prediction with Current Clinical Blood Reservation Practice

  Perioperative (from the start of surgery to 24 hours postoperatively)

• 30-Day In-Hospital Mortality

  30 days after surgery

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Patients aged over 18 years with ASA physical status I-III who underwent oncologic surgery at a single center between April 3, 2023, and the date of ethics committee approval, identified retrospectively from hospital medical records.

You may qualify if:

• Patients over 18 years of age
• Underwent oncologic surgery
• ASA physical status I-III

You may not qualify if:

• Underwent emergency surgery
• Hematologic disease
• Pregnancy
• Incomplete data
• Underwent cardiac surgery

Sponsors & Collaborators

• Kocaeli City Hospital: lead

Study Sites (1)

Kocaeli City Hospital

Kocaeli, Turkey (Türkiye)

Location

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER GOV
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: September 25, 2026
• First Posted: October 15, 2026
• Study Start: August 9, 2025
• Primary Completion (Estimated): December 1, 2026
• Study Completion (Estimated): December 31, 2026
• Last Updated: October 15, 2026

Record last verified: 2026-10

Data Sharing

• IPD Sharing: Will not share

Locations

TU (1)