Standalone Model Search Spaces for Clinical Disease Prediction

A 5.5 KB Excel file containing a unified deep learning framework evaluated on three benchmark clinical datasets. The framework, proposed by Vijay U. Rathod, integrates MLP, CNN, FT-Transformer, and autoencoder architectures. It was last updated on May 8, 2026, and demonstrated competitive performance, with an FT-Transformer + autoencoder ensemble achieving an AUC of 0.8980 for heart disease prediction.

Use Cases

• Benchmarking deep learning models for heart disease prediction based on 13 clinical features from the UCI dataset
• Evaluating model performance on imbalanced datasets for diabetes classification based on 8 metabolic features from the PIMA dataset
• Developing ensemble strategies for Parkinson's disease detection based on 22 acoustic features from voice recordings

Strengths

• Framework evaluated on three distinct benchmark datasets with specified sample sizes: UCI Heart Disease (303 samples), PIMA Diabetes (768 samples), and Parkinson's voice (195 recordings)
• Reports specific model performance metrics, including an AUC of 0.8451 for diabetes classification and perfect specificity for Parkinson's disease detection with an MLP
• Employs robust evaluation methods including nested cross-validation and fold-safe feature selection as described

Limitations

• Row count and column-level documentation are unknown, which limits suitability assessment and requires manual inspection after download
• The dataset is very small at 5.5 KB, indicating limited scope, likely containing framework specifications or results rather than raw clinical data

Provenance

Source

Vijay U. Rathod via figshare

Collection Method

Likely contains the proposed framework's model search spaces or evaluation results, as described in the accompanying study.

Freshness

Last updated 2026-05-08 17:34:22; freshness should be verified