QuanvNeXt: An end-to-end quanvolutional neural network for EEG-based detection of major depressive disorder

TL;DR

QuanvNeXt is a novel end-to-end quanvolutional neural network that effectively diagnoses major depressive disorder from EEG data, outperforming existing models in accuracy and reliability, with strong interpretability and calibration.

Contribution

This paper introduces QuanvNeXt, a new quanvolutional neural network with a Cross Residual block, enhancing feature learning and efficiency for EEG-based depression detection.

Findings

Achieved 93.1% accuracy and 97.2% AUC-ROC on two datasets.

Outperformed state-of-the-art models like InceptionTime.

Demonstrated well-calibrated predictions under noise perturbations.

Abstract

This study presents QuanvNeXt, an end-to-end fully quanvolutional model for EEG-based depression diagnosis. QuanvNeXt incorporates a novel Cross Residual block, which reduces feature homogeneity and strengthens cross-feature relationships while retaining parameter efficiency. We evaluated QuanvNeXt on two open-source datasets, where it achieved an average accuracy of 93.1% and an average AUC-ROC of 97.2%, outperforming state-of-the-art baselines such as InceptionTime (91.7% accuracy, 95.9% AUC-ROC). An uncertainty analysis across Gaussian noise levels demonstrated well-calibrated predictions, with ECE scores remaining low (0.0436, Dataset 1) to moderate (0.1159, Dataset 2) even at the highest perturbation ({\epsilon} = 0.1). Additionally, a post-hoc explainable AI analysis confirmed that QuanvNeXt effectively identifies and learns spectrotemporal patterns that distinguish between healthy controls and major depressive disorder. Overall, QuanvNeXt establishes an efficient and reliable approach for EEG-based depression diagnosis.