Table 26 Deployment Configuration and Inference Performance of NeuroFusionNet.

From: NeuroFusionNet: a hybrid EEG feature fusion framework for accurate and explainable Alzheimer’s Disease detection

Deployment parameter

Specification / result

Hardware Platform

Intel Core i7 (2.8 GHz), 16 GB RAM

GPU Acceleration

Not Used (CPU-Only Inference)

Model Parameters

0.94 Million

Model Footprint

4.1 MB

Average Inference Time

6.5 ms per sample (mean ± SD = 6.5 ± 0.3 ms)

Throughput

\(\sim\)150 samples per second

Real-Time Capability

Achieved (Yes)

Deployment Targets

Portable EEG devices / Bedside monitors

Back to article page