Clinical Disease Prediction Benchmark Datasets for Deep Learning Framework Evaluation

Three benchmark datasets for evaluating a unified deep learning framework for disease prediction. The data includes 303 samples for heart disease, 768 for diabetes, and 195 voice recordings for Parkinson's disease. Author Vijay U. Rathod published the dataset on figshare in 2026.

Use Cases

• Benchmarking deep learning models for heart disease prediction based on 13 clinical features
• Evaluating model performance on diabetes classification based on 8 metabolic features
• Testing framework generalization on Parkinson's disease detection based on 22 acoustic features
• Comparing ensemble strategies like FT-Transformer + autoencoder for clinical tasks

Strengths

• Includes three distinct clinical datasets with sample sizes of 303, 768, and 195
• Reports specific model performance metrics, such as an AUC of 0.8980 for heart disease prediction
• Framework employs nested cross-validation and fold-safe feature selection for reliable evaluation

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown for the aggregated file, which may limit suitability assessment
• Data may reflect bias inherent to the specific benchmark datasets used

Provenance

Source

Benchmark datasets from UCI and other sources, aggregated by the author.

Collection Method

Likely compiled from public repositories for framework evaluation.

Freshness

Last updated 2026-05-08 17:34:28; freshness should be verified