Predictive Model for Breast Cancer Lymphedema Surgery Outcomes, Beijing 2018-2025

A retrospective cohort study of 300 patients from Beijing Shijitan Hospital between June 2018 and June 2025. The dataset likely contains clinical variables used to develop and validate a machine learning model predicting 6-month outcomes after combined suction-assisted lipectomy and lymphovenous anastomosis surgery. The predictive model, authored by Yonghao Cui, was published on figshare in May 2026.

Use Cases

• Predicting 6-month surgical outcomes based on postoperative excess limb volume, disease duration, and severity grade.
• Comparing performance of seven machine learning algorithms (LR, DT, RF, XGBoost, LightGBM, SVM, ANN) for clinical risk stratification.
• Developing web-based clinical decision support tools for early postoperative monitoring of high-risk patients.

Strengths

• 300 patient records with a 72.3% rate of satisfactory outcomes at 6 months.
• Model validation includes AUC, calibration plots, decision curve analysis, and Brier score.
• Three stable predictors identified via LASSO regression with bootstrap validation (selection frequencies: 100%, 83.6%, 83.4%).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is from a single hospital in Beijing and may reflect geographic bias.

Provenance

Source

Beijing Shijitan Hospital

Collection Method

Retrospective cohort study

Time Range

June 2018 to June 2025

Freshness

Last updated 2026-05-07 04:44:57

Geography

Beijing, China