AI-assisted Decision-making of Reoperation for Postoperative Bleeding of Gastric Cancer

Sponsor
First Affiliated Hospital of Zhejiang University
Study ID
NCT07525765
Status
Recruiting
Apply to this trial

Conditions

Eligibility Criteria

Sex
ALL
Age
18 Years - 90 Years
Healthy Volunteers
Not accepted

Study Details

The goal of this observational study is to develop and validate a deep learning model to dynamically assess postoperative bleeding risk and assist in decision-making for re-operation in adult patients (≥18 years) diagnosed with primary gastric cancer undergoing radical gastrectomy. The main question\[s\] it aims to answer \[is/are\]: Can an AI model based on perioperative dynamic physiological parameters and precise intraoperative blood loss accurately predict the risk of postoperative bleeding requiring re-operation? Does the application of this AI model improve clinical decision-making (e.g., earlier warning time, optimal intervention timing) and patient outcomes (e.g., mortality, length of stay)? Since there is no comparison group (this is a pure observational study without intervention arms), researchers will not compare different treatment groups. Instead, the investigators will evaluate the model's performance (sensitivity, negative predictive value, AUC, calibration) using retrospective data for training and prospective multi-center data for external validation. Participants will: Undergo standard radical gastrectomy and routine postoperative care as per clinical practice (no study-specific interventions). Have their perioperative data collected, including demographics, medical history, vital signs, laboratory tests (blood gas analysis), surgical details, and precise intraoperative blood loss measurements. (For prospective participants only) Provide informed consent and complete follow-up assessments up to 30 days post-surgery.

Key Dates

First listed
Apr 13, 2026
Start date
Apr 10, 2026
Status verified
Mar 2026
Primary completion
Dec 31, 2027
Completion
Jan 31, 2028

Study Design

Enrollment
7,000 participants (estimated)

Arms

  • Arm: Training set (led by the Principal Investigator)
    The main part of retrospective data for model construction, parameter learning, without interventions
  • Arm: Validation set (led by the Principal Investigator)
    The remainder of the retrospective data for hyperparameter tuning to prevent overfitting, without interventions
  • Arm: External validation set (conducted by other investigators)
    Prospective collected data for final performance evaluation, without interventions

Primary Outcome Measure

predictive performance of the deep learning model for identifying patients at high risk of postoperative bleeding requiring re-operation [ Time Frame: The primary endpoint is the AUC-ROC of the model in predicting postoperative bleeding requiring re-operation within 30 days after surgery ]

Central Contacts

Find similar trials

Nationwide
Gastric cancer trialsOncology trialsGI oncology trials

Related Studies