<p>Quality measure after denoising.</p>

A 5.5 KB Excel file containing quality metrics from a proposed deep learning framework for diagnosing Alzheimer's disease (AD), mild cognitive impairment (MCI), and healthy subjects (HSs) using electroencephalography (EEG) data. The framework, called the Cognitive Decline Recognition Network (CDR-Net), achieved reported multiclass accuracy, sensitivity, and specificity of 99.25%, 99.13%, and 99.32%, respectively. The dataset was authored by Ashik Mostafa Alvi and last updated on April 20, -2026.

Use Cases

• Benchmarking new EEG classification models against the reported 99.25% accuracy metric.
• Analyzing the performance stability of diagnostic frameworks using the described 10-fold and leave-one-out cross-validations.
• Investigating feature extraction and preprocessing methods for EEG data as outlined in the proposed framework.

Strengths

• Reported high-performance metrics: 99.25% accuracy, 99.13% sensitivity, and 99.32% specificity for multiclass classification.
• Framework addresses data overfitting and underfitting concerns through cross-validation methods mentioned in the description.
• Data is shared under a permissive CC-BY-4.0 license.

Limitations

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

Provenance

Source

Ashik Mostafa Alvi via figshare

Collection Method

Likely derived from research proposing the CDR-Net architecture for EEG-based diagnosis.

Time Range

null

Freshness

Last updated 2026-04-20 17:58:07; freshness should be verified.

Geography

null