Signal Classification Results: Needle EMG Data for Neuromuscular Disease Diagnosis

376 needle electromyography (nEMG) signals from 57 patients were used to validate a deep learning diagnosis-aiding system. The model classified signals into normal, myopathy, and neuropathy categories and surpassed the performance of six electromyographers. Ilhan Yoo published the results on figshare in May 2026.

Use Cases

• Benchmarking AI classification models for medical signals based on the three diagnostic categories
• Analyzing model interpretability and failure cases based on the reported signal features
• Comparing automated diagnostic performance against human experts based on the reported accuracy metrics

Strengths

• 376 nEMG signals provide a concrete sample size for analysis
• Performance metrics (accuracy, precision, sensitivity, specificity) for both the model and physicians are reported
• Data is validated using nested k-fold cross-validation

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• The dataset is very small (23.5 KB), indicating limited scope

Provenance

Source

A tertiary university hospital database

Collection Method

Data was collected and used to validate a deep learning model with minimal preprocessing

Freshness

Last updated 2026-05-19 17:38:37