Clinical Data for Ankylosing Spondylitis Autoimmune Comorbidity Analysis, 2012-2025

A multicenter, retrospective, cross-sectional study collected clinical and laboratory data from three independent tertiary centers between 2012 and 2025. The dataset includes demographic characteristics, systemic inflammatory indices, hematological parameters, and liver and renal function markers for patients with ankylosing spondylitis alone, autoimmune diseases alone, or AS with autoimmune comorbidities. It was authored by Jichong Zhu and shared under a CC-BY-4.0 license.

Use Cases

• Train classification models to distinguish AS alone from AS with autoimmune comorbidities based on demographic and renal function markers.
• Develop models to differentiate autoimmune diseases alone from AS with autoimmune comorbidities using metabolic and inflammatory features.
• Apply SHAP analysis for model interpretability to identify key clinical contributors like age, gender, eGFR, and ESR.
• Benchmark machine learning algorithms like LightGBM for performance on clinical discrimination tasks.

Strengths

• Data spans a 13-year period from 2012 to 2025, providing a longitudinal clinical perspective.
• Includes a variety of routinely available clinical variables such as demographic, inflammatory, hematological, and organ function markers.
• Model performance was evaluated using multiple metrics including AUC, calibration, and clinical impact curves.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 11.6 KB, indicating limited scope or a summary table rather than raw patient records.

Provenance

Source

figshare, author Jichong Zhu

Collection Method

Multicenter, retrospective, cross-sectional study collecting clinical and laboratory data.

Time Range

2012 to 2025

Freshness

Last updated 2026-03-18 07:40:05

Geography

Data collected from three independent tertiary centers; specific locations unknown.