DC-AKI: Continuous Acute Kidney Injury Risk Predictions for Critically Ill Patients

A 15.5 MB dataset supporting the development and validation of a deep learning model for real-time acute kidney injury prediction. The data, authored by SIYU TANG and last updated on 2026-05-28, is shared under a CC-BY-4.0 license on figshare. It likely contains patient-level features used to predict 48-hour AKI risk in intensive care settings.

Use Cases

• Training a deep recurrent neural network for continuous AKI risk prediction based on patient monitoring data.
• Benchmarking clinical prediction models against the described DC-AKI architecture.
• Studying feature importance for 48-hour AKI onset in critically ill populations.

Strengths

• Dataset is 15.5 MB, indicating a defined and downloadable resource.
• Model validation is explicitly mentioned, suggesting a structured experimental setup.
• Data is shared under a permissive CC-BY-4.0 license, allowing for reuse and modification.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale model training.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely derived from electronic health records or critical care databases for model development.

Freshness

Last updated 2026-05-28 11:43:02; freshness should be verified.