Patient Classification Results from a Needle Electromyography Diagnosis-Aiding System

A 2026 study by Ilhan Yoo validated a deep learning model for classifying patients based on needle electromyography (nEMG) signals. The dataset contains results from 376 nEMG signals collected from 57 patients at a tertiary university hospital. The model's performance was compared against six electromyographers using nested k-fold cross-validation.

Use Cases

• Benchmarking machine learning models for medical signal classification based on nEMG data.
• Studying model interpretability and failure analysis in clinical AI based on signal features.
• Comparing AI-assisted diagnostic performance against human experts in neuromuscular disease assessment.

Strengths

• Includes performance metrics (accuracy, precision, sensitivity, specificity) for both a deep learning model and six physicians.
• Based on 376 nEMG signals from 57 patients, providing a concrete scale for validation.
• Uses a nested k-fold cross-validation methodology, which suggests a rigorous evaluation approach.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The dataset is very small at 7.7 KB, indicating limited scope, likely containing summary results rather than raw signal data.
• Data may reflect bias inherent to the single tertiary university hospital source.

Provenance

Source

Tertiary university hospital database, author Ilhan Yoo.

Collection Method

nEMG signals were collected and used to train and validate a deep learning model via nested k-fold cross-validation.

Freshness

Last updated 2026-05-19 17:38:38