NeuroHAGWO-Net Model Performance Metrics for Mental Health Detection

NeuroHAGWO-Net, an AI framework for mental health detection, achieved an accuracy of 0.9945 on behavioral data and 0.9935 on EEG signals. The dataset, authored by Sunil Kumar Sharma and last updated in April 2026, contains performance metrics for this multimodal model combining EEG and text data.

Use Cases

• Benchmarking new multimodal AI models for mental health detection based on the reported accuracy, precision, and F1-score metrics.
• Analyzing the efficacy of Hybrid Ant-Grey Wolf Optimization (HAGWO) for feature selection in clinical AI applications.
• Comparing performance of deep learning architectures like BiLSTM-CNN and NeuroVisionNet on EEG and text data modalities.
• Evaluating the trade-offs between sensitivity, specificity, and false positive/negative rates in automated diagnostic tools.

Strengths

• Performance metrics are highly detailed, including 9 distinct measures such as MCC (0.9925) and NPV (0.9905).
• Model validation was performed on two distinct data modalities: behavioral text and EEG signals.
• Dataset is openly licensed under CC-BY-4.0, facilitating reuse and sharing.

Limitations

• Dataset is very small at 5.5 KB, suggesting it contains only summary performance statistics, not the underlying raw data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for training or validation tasks.

Provenance

Source

figshare, author Sunil Kumar Sharma

Collection Method

Performance metrics from validating the NeuroHAGWO-Net AI model on two datasets.

Time Range

Publication date suggests model development circa 2026.

Freshness

Last updated 2026-04-13 17:44:11; freshness should be verified.

Geography

null