An Explainable Deep Learning-Based Method For Schizophrenia Diagnosis Using Generative Data-Augmentation

TL;DR

This paper presents a deep learning approach for schizophrenia diagnosis using EEG data, enhanced by generative data augmentation with VAE and WGAN-GP, and interpretable with LIME, achieving high accuracy and transparency.

Contribution

It introduces a novel combination of CNN with generative models for data augmentation and applies LIME for interpretability in schizophrenia diagnosis.

Findings

VAE-augmented dataset improved accuracy to 99%.

Generative augmentation reduced overfitting and convergence time.

LIME identified key spectral features for diagnosis.

Abstract

In this study, we leverage a deep learning-based method for the automatic diagnosis of schizophrenia using EEG brain recordings. This approach utilizes generative data augmentation, a powerful technique that enhances the accuracy of the diagnosis. To enable the utilization of time-frequency features, spectrograms were extracted from the raw signals. After exploring several neural network architectural setups, a proper convolutional neural network (CNN) was used for the initial diagnosis. Subsequently, using Wasserstein GAN with Gradient Penalty (WGAN-GP) and Variational Autoencoder (VAE), two different synthetic datasets were generated in order to augment the initial dataset and address the over-fitting issue. The augmented dataset using VAE achieved a 3.0\% improvement in accuracy reaching up to 99.0\% and yielded a lower loss value as well as a faster convergence. Finally, we addressed the lack of trust in black-box models using the Local Interpretable Model-agnostic Explanations (LIME) algorithm to determine the most important superpixels (frequencies) in the diagnosis process.