On the Validity of Head Motion Patterns as Generalisable Depression Biomarkers

TL;DR

This study investigates the use of head motion patterns, called kinemes, as reliable and generalisable biomarkers for depression severity assessment across diverse datasets and cultural contexts, demonstrating their effectiveness over other behavioral cues.

Contribution

It introduces the concept of kinemes for depression analysis and evaluates their generalisability across multiple datasets and cultural settings, highlighting their potential as robust biomarkers.

Findings

Head motion patterns are effective biomarkers for depression severity estimation.

Kineme-based features outperform raw motion descriptors and other visual cues in generalisability.

Models achieved competitive performance, including second-best MAE on AVEC2013.

Abstract

Depression is a debilitating mood disorder negatively impacting millions worldwide. While researchers have explored multiple verbal and non-verbal behavioural cues for automated depression assessment, head motion has received little attention thus far. Further, the common practice of validating machine learning models via a single dataset can limit model generalisability. This work examines the effectiveness and generalisability of models utilising elementary head motion units, termed kinemes, for depression severity estimation. Specifically, we consider three depression datasets from different western cultures (German: AVEC2013, Australian: Blackdog and American: Pitt datasets) with varied contextual and recording settings to investigate the generalisability of the derived kineme patterns via two methods: (i) k-fold cross-validation over individual/multiple datasets, and (ii) model reuse on other datasets. Evaluating classification and regression performance with classical machine learning methods, our results show that: (1) head motion patterns are efficient biomarkers for estimating depression severity, achieving highly competitive performance for both classification and regression tasks on a variety of datasets, including achieving the second best Mean Absolute Error (MAE) on the AVEC2013 dataset, and (2) kineme-based features are more generalisable than (a) raw head motion descriptors for binary severity classification, and (b) other visual behavioural cues for severity estimation (regression).