Thoracic Surgery Data: Post-Operative Survival Prediction for Lung Cancer Patients

470 patient records from the Wroclaw Thoracic Surgery Centre for individuals who underwent major lung resections for primary lung cancer between 2007 and 2011. The dataset is dedicated to predicting one-year post-operative survival, with a binary target variable indicating death or survival. It was collected by Marek Lubicz et al. as part of the National Lung Cancer Registry in Poland.

Use Cases

• Predict one-year survival after lung cancer surgery based on patient diagnosis, lung function tests, and symptoms.
• Analyze the impact of pre-operative conditions like pain, cough, or diabetes on post-operative mortality.
• Build classification models using clinical features such as FVC, FEV1, performance status, and tumour size.
• Study risk factor correlations, such as smoking or asthma, with surgical outcomes in a specific patient cohort.

Strengths

• 470 patient instances with a clear binary classification target (70 deaths, 400 survivors).
• Contains 17 clinical and demographic attributes, including numeric lung function measures and binary symptom indicators.
• Data sourced from a recognized medical centre and integrated into a national cancer registry, suggesting a formal collection process.

Limitations

• Row count and column details beyond the listed features are unknown, limiting suitability assessment.
• Data is from a single centre in Poland (2007-2011), which may limit geographic and temporal generalizability.
• The last update date is unknown, so data freshness is unverified.

Provenance

Source

Wroclaw Thoracic Surgery Centre, part of the National Lung Cancer Registry administered by the Institute of Tuberculosis and Pulmonary Diseases in Warsaw, Poland.

Collection Method

Retrospectively collected for patients undergoing major lung resections for primary lung cancer.

Time Range

2007 to 2011

Freshness

null

Geography

Wroclaw, Poland