Machine Learning to Predict Postoperative Pneumonia in Brain Tumor Patients (ML-PNEUMO-BT)

Ming Yang

Status: Completed

Conditions: Brain Tumors; Postoperative Pneumonia

Study type: Observational

Funder types: Other

Identifiers

NCT07321262

NCC2025C-1426

2024-I2M-3-014 (Other Identifier)

Details and patient eligibility

About

Postoperative pneumonia (POP) is a common and serious complication after elective craniotomy for brain tumor resection. POP often develops within the first week after surgery and may lead to prolonged hospitalization, higher medical costs, and increased risk of severe illness. Because symptoms can be subtle in neurosurgical patients, POP may be detected late, limiting timely prevention and treatment.

This study will evaluate whether a machine-learning-based clinical decision support tool can help clinicians identify patients at high risk for POP early and improve perioperative preventive care. The tool uses routinely collected clinical information to estimate an individual patient's POP risk and provides an easy-to-understand explanation of key risk drivers. Based on the predicted risk level (low, moderate, high, or very high), the system suggests standardized preventive care pathways (e.g., perioperative airway management, targeted antibiotic strategies per local practice, and nutritional support), while allowing clinicians to override recommendations at any time.

Participants will be adults undergoing their first elective craniotomy for brain tumor resection at participating neurosurgical centers. The primary outcome is the occurrence of POP within 7 days after surgery, defined using CDC/NHSN criteria. Secondary outcomes include antibiotic use intensity, length of hospital stay, direct medical cost, and clinician decision confidence. Participants will be followed at postoperative days 1, 3, and 7 using electronic medical record review and phone confirmation when needed.

The goal of this study is to determine whether integrating an explainable AI risk prediction tool into routine care can reduce POP and improve the quality and efficiency of perioperative management after brain tumor surgery.

Full description

Rationale Postoperative pneumonia (POP) remains a frequent and clinically important complication after elective craniotomy for brain tumor resection, contributing to prolonged hospitalization, increased cost, and worse clinical outcomes. Conventional POP risk assessment is often experience-based or relies on simplified scoring approaches, which may not adequately capture nonlinear interactions among perioperative factors. This study implements an explainable machine-learning (ML) prediction model within routine perioperative workflows and evaluates whether model-assisted care can improve POP prevention and related resource utilization compared with usual care.

Decision support system

An explainable gradient boosting machine (GBM) model is used to estimate individual POP risk from routinely available perioperative variables. Interpretability is provided using SHAP-based explanations at two complementary levels:

Population level: summarizes the most influential predictors and selected interaction patterns to support clinical understanding and model governance.

Patient level: generates an individualized contribution visualization (e.g., waterfall-style), highlighting the main drivers of a specific patient's risk estimate.

The system automatically assigns a risk tier (low, moderate, high, or very high) and links each tier to standardized prevention pathway templates (e.g., airway management optimization, antimicrobial stewardship-consistent strategies per local policy, and nutritional support). The tool does not mandate treatment; clinicians may accept, modify, or override any suggestion.

Evaluation framework The overall project includes retrospective model development/optimization, prospective external validation/calibration, and a pragmatic implementation evaluation. The registered interventional evaluation uses a multicenter, cluster randomized crossover design with monthly alternating periods of model-assisted care versus usual care. Allocation procedures, eligibility criteria, planned enrollment, and endpoint definitions/time windows are specified in the corresponding record modules (Study Design, Arms/Interventions, Outcome Measures, and Eligibility) to avoid duplication in this section.

Implementation and integration The model is deployed as a lightweight web service with unified APIs and data-exchange formats to enable non-disruptive integration with hospital information systems (HIS) and electronic medical records (EMR). A web-based front end and a Python-based back end support RESTful calls and are designed for low-latency inference (target single-prediction latency <200 ms), suitable for perioperative and inpatient workflows.

Data governance and model updating To support long-term generalizability across hospitals and mitigate dataset shift, the project establishes a closed-loop maintenance process ("local de-identification → cloud retraining → model version management → edge deployment"). Model updates are version-controlled and deployed under governance procedures consistent with local regulations, institutional policies, and applicable ethics approvals.

Enrollment: 1,856 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Age ≥ 18 years.
• Undergoing elective craniotomy for intracranial brain tumor resection.
• Perioperative clinical data available in the electronic medical record to derive required predictors.
• Expected postoperative survival ≥ 7 days.

Exclusion criteria

• Evidence of active infection (including pneumonia) prior to surgery.
• Thoracic surgery or severe chest trauma within 30 days prior to craniotomy.
• Spinal tumors or extracranial peripheral nerve tumors.
• Pregnancy or lactation.
• Hospice care, expected survival < 7 days, or insufficient data completeness for model calculation.

Trial design

1,856 participants in 5 patient groups

Retrospective model development cohort (CAMS/PUMC, 2022-2024)

Description:

Retrospective cohort collected at the Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College (CAMS/PUMC) between Jan 1, 2022 and Oct 31, 2024. Adult patients undergoing elective craniotomy for intracranial brain tumor resection were included. This cohort was used for model development, comparison of candidate algorithms, and internal validation; it was randomly split 8:2 into a training set (n=609) and an internal validation set (n=152). Early postoperative pneumonia (POP) was defined by CDC criteria within 7 days postoperatively. No study-mandated intervention was applied; routine perioperative care was provided.

Prospective internal validation cohort (Test P, CAMS/PUMC)

Description:

Prospective cohort enrolled at CAMS/PUMC between Nov 1, 2024 and Apr 30, 2025 (Test P; n=224). Adult patients undergoing elective craniotomy for brain tumor resection were included. This cohort prospectively validated the finalized interpretable prediction model for early POP using routinely available perioperative EMR variables. Early POP was defined per CDC criteria within 7 postoperative days. Clinical management followed standard-of-care without any study-assigned intervention.

Prospective external validation cohort (Test A, Anhui Medical Univ)

Description:

Prospective external validation cohort recruited at the First Affiliated Hospital of Anhui Medical University from Aug 1, 2024 to Apr 30, 2025 (Test A; n=329). Adult patients undergoing elective craniotomy for brain tumor resection were included. The cohort independently validated the finalized POP prediction model using routine perioperative data. Early POP was defined by CDC criteria within 7 days after surgery. No investigational intervention was administered; all care followed local standard practice.

Prospective external validation cohort (Test S, Shandong Cancer Hospital)

Description:

Prospective external validation cohort recruited at Shandong Cancer Hospital from Aug 1, 2024 to Apr 30, 2025 (Test S; n=440). Adult patients undergoing elective craniotomy for intracranial brain tumor resection were included. This cohort was used to externally validate the finalized interpretable model for early POP prediction using routinely collected perioperative variables. Early POP was defined according to CDC criteria within 7 postoperative days. Participants received routine perioperative management per institutional standards without study-mandated interventions.

Prospective external validation cohort (Test J, Jinan Fourth People's Hospital)

Description:

Prospective external validation cohort recruited at Jinan Fourth People's Hospital from Aug 1, 2024 to Apr 30, 2025 (Test J; n=102). Adult patients undergoing elective craniotomy for brain tumor resection were included. The cohort externally validated the finalized prediction model for early POP based on routine perioperative EMR data. Early POP was defined per CDC criteria within 7 postoperative days. There was no study-assigned intervention; all patients received standard perioperative care per local protocols.