Data Sheet 1_Quantitative peri-lesional densitometry mapping via thin-slice volume renderi

A retrospective study of 383 patients (193 with pneumonic-type lung cancer, 190 with inflammatory pneumonia) for model development and 61 patients for external validation. The dataset, authored by Sifan Chen and last updated in April 2026, supports a novel diagnostic model integrating densitometry-derived thresholds into thin-slab volume rendering (tsVR) to differentiate these conditions.

Use Cases

• Train a binary classification model to differentiate pneumonic-type lung cancer from inflammatory pneumonia based on peri-lesional densitometric analysis.
• Benchmark diagnostic performance of a thin-slab volume rendering (tsVR) model against conventional CT feature-based diagnosis using metrics like AUC, sensitivity, and specificity.
• Conduct logistical regression analyses to assess independent factors influencing diagnostic accuracy across different scanners and protocols.

Strengths

• Includes 383 patients for model development and 61 for external validation, providing a basis for internal and external testing.
• The tsVR model demonstrated an AUC of 0.86 in internal validation and 0.81 in external validation, with reported sensitivity of 0.90.
• Released under a CC-BY-4.0 license, permitting broad reuse with attribution.

Limitations

• Row count and column-level documentation are unknown, which may limit suitability assessment and require manual inspection after download.
• The dataset is very small at 84.9 KB, indicating limited scope, likely containing summary results or a document rather than raw patient-level data.
• The data is retrospective and may reflect biases inherent to the specific patient cohorts and imaging protocols used.

Provenance

Source

figshare, authored by Sifan Chen.

Collection Method

Retrospective study using patient CT scans for densitometric analysis and model development.

Time Range

null

Freshness

Last updated 2026-04-28 04:24:57; freshness should be verified.

Geography

null