Predicting Depression Severity by Multi-Modal Feature Engineering and Fusion
This work addresses depression assessment for clinical applications, but it appears incremental as it builds on existing multi-modal fusion approaches.
The paper tackled predicting depression severity using vocal, linguistic, and facial patterns, achieving results that outperformed single-modality models and the dataset baseline.
We present our preliminary work to determine if patient's vocal acoustic, linguistic, and facial patterns could predict clinical ratings of depression severity, namely Patient Health Questionnaire depression scale (PHQ-8). We proposed a multi modal fusion model that combines three different modalities: audio, video , and text features. By training over AVEC 2017 data set, our proposed model outperforms each single modality prediction model, and surpasses the data set baseline with ice margin.