Fertility Intentions in Crohn's Disease Patients with Psychosocial and Clinical Variables

Jinghan Liu's dataset contains records from 276 Crohn's disease patients aged 18–45, prospectively recruited from a tertiary center in China. The data was used to construct and validate interpretable machine learning models for predicting fertility intentions. The dataset was last updated on May 22, 2026.

Use Cases

• Training predictive models for fertility intentions based on demographic, clinical, and psychosocial variables mentioned in the description.
• Identifying key psychosocial determinants of reproductive decision-making using SHAP analysis.
• Benchmarking machine learning algorithms like neural networks and XGBoost on a clinical prediction task.
• Conducting secondary analysis on the relationship between disease activity, marital status, and family support on fertility plans.

Strengths

• Includes data from 276 prospectively recruited patients, providing a defined cohort size.
• Model performance metrics are reported, with neural networks achieving an AUC of 0.718.
• Key influential predictors like marital status and perceived family support are identified through SHAP analysis.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data reflects a single-center study in China, which may limit generalizability to other populations.

Provenance

Source

Jinghan Liu via figshare.

Collection Method

Prospective recruitment and standardized assessments of patients from a tertiary center.

Freshness

Last updated 2026-05-22 06:10:56; freshness should be verified.

Geography

China (single tertiary center).