A Multimodal Framework for the Assessment of the Schizophrenia Spectrum

TL;DR

This paper introduces a multimodal framework combining audio, video, and text data with novel fusion techniques to improve classification of schizophrenia spectrum symptoms versus healthy controls.

Contribution

The paper proposes a new multimodal fusion approach using minimal Gated multimodal units (mGMU) for better symptom classification in schizophrenia spectrum disorders.

Findings

Improved weighted F1-score with mGMU fusion

Enhanced weighted AUC-ROC scores using the proposed framework

Effective integration of audio, video, and text modalities

Abstract

This paper presents a novel multimodal framework to distinguish between different symptom classes of subjects in the schizophrenia spectrum and healthy controls using audio, video, and text modalities. We implemented Convolution Neural Network and Long Short Term Memory based unimodal models and experimented on various multimodal fusion approaches to come up with the proposed framework. We utilized a minimal Gated multimodal unit (mGMU) to obtain a bi-modal intermediate fusion of the features extracted from the input modalities before finally fusing the outputs of the bimodal fusions to perform subject-wise classifications. The use of mGMU units in the multimodal framework improved the performance in both weighted f1-score and weighted AUC-ROC scores.