Data Sheet 1_Development and validation of a multimodal clinical-radiomics-deep learning n

Qiang Fei authored a dataset supporting a multicenter study to classify Chronic Obstructive Pulmonary Disease (COPD) severity. The data includes clinical characteristics, radiomic features, and deep learning features from automated chest CT segmentation for 1,794 patients across three hospitals. The study period spans from January 1, 2021, to May 30, 2025.

Use Cases

• Training machine learning models to classify COPD severity based on clinical and imaging features.
• Benchmarking radiomics models against deep learning models for medical image analysis tasks.
• Developing multimodal predictive tools for patient management and treatment planning in respiratory disorders.

Strengths

• Data originates from a multicenter study involving 1,794 patients, which likely improves generalizability.
• Model performance is validated across training, internal validation, and external validation cohorts.
• The combined model achieved an AUC of up to 0.839, outperforming individual clinical, radiomics, and deep learning models.

Limitations

• The dataset is described in a 20.1 KB DOCX file, suggesting limited raw data or that it primarily contains summary results and methodology.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for direct model training.

Provenance

Source

figshare

Collection Method

Retrospective analysis of patient data from three hospitals.

Time Range

January 1, 2021 to May 30, 2025

Freshness

Last updated 2026-05-13 05:33:58; freshness should be verified.