Cardiac Surgery Patient Sleep Disturbance Risk Model Data

Lin Jierong's dataset supports models for predicting postoperative sleep disturbance in cardiac surgery patients. The data originates from a prospective study conducted between November 2023 and May 2024 and a subsequent randomized controlled trial from November 2024 to January 2025. The dataset is 14.2 KB in size and is stored in SAV format.

Use Cases

• Predict PSD risk using logistic regression models based on patient features like surgical approach (non-thoracotomy vs. thoracotomy).
• Build a personalized advantage index (PAI) model using random forest and L1-regularized regression on identified risk factors.
• Compare PSD incidence and PSQI scores between intervention and control groups from the 70-patient RCT data.
• Validate model performance using reported metrics such as AUC, sensitivity, and specificity.

Strengths

• Data supports separate validated models for non-thoracotomy and thoracotomy patient subgroups.
• Includes results from a randomized controlled trial with 70 patients, providing intervention outcome data.

Limitations

• The dataset is very small at 14.2 KB, indicating limited raw data or a summary of model results.
• Specific column names, row counts, and individual patient-level features are not provided in the metadata.
• Geographic origin and patient demographic details are unknown, limiting generalizability.

Provenance

Source

Lin Jierong via figshare.

Collection Method

Prospective observational study followed by a stratified randomized controlled trial.

Time Range

Primary study: November 2023 to May 2024. RCT: November 2024 to January 2025.

Freshness

Data collection concluded in January 2025, and the record was last updated in April 2026.