Hormonal and Laboratory Predictors of Patent Foramen Ovale in 300 Female Patients

300 female patients aged 18–65 with cryptogenic ischemic events were analyzed for predictors of patent foramen ovale. The study by Yao Zheng, published in 2026, identified five key predictors including age, estradiol, and D-dimer. A logistic regression model built on these features achieved an AUC of 0.990 and 95.6% accuracy.

Use Cases

• Developing predictive models for PFO detection based on clinical and laboratory variables like estradiol and D-dimer.
• Analyzing feature importance for stroke risk using SHAP analysis on hormonal and lipid markers.
• Conducting subgroup analysis of medical conditions by reproductive stage (reproductive-age vs. menopausal).

Strengths

• Includes 300 patient records with 43 analyzed variables.
• Model performance metrics are high, with an AUC of 0.990 and accuracy of 95.6%.
• Features were selected using LASSO and mRMR methods, and model interpretability was enhanced with SHAP analysis.

Limitations

• Row count for the underlying dataset is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is limited to 300 female patients from a single retrospective study.

Provenance

Source

Yao Zheng via figshare

Collection Method

Retrospective study of patients who underwent contrast-enhanced transcranial Doppler and echocardiography.

Freshness

Last updated 2026-04-17 04:12:16; freshness should be verified.