On the Validity of Head Motion Patterns as Generalisable Depression Biomarkers
Monika Gahalawat, Maneesh Bilalpur, Raul Fernandez Rojas, Jeffrey F. Cohn, Roland Goecke, Ramanathan Subramanian
TL;DR
This work addresses the generalisability of depression biomarkers by introducing kineme-based head-motion features learned from healthy controls and applying them to three cross-cultural datasets (AVEC2013, Pitt, Blackdog) for depression severity estimation. It combines segmentation, non-negative matrix factorisation and Gaussian mixture modelling to derive 16 kinemes, then uses reconstruction-error–derived, chunk-level statistical features for classification and regression. Across two evaluation paradigms—k-fold cross-validation and cross-dataset transfer—the kineme features demonstrate strong generalisability, outperforming raw head pose and facial cues in several configurations and achieving competitive MAE/RMSE on AVEC2013. The findings suggest kineme patterns offer robust, culture-agnostic biomarkers with practical impact for automated, objective depression assessment and point to future multimodal extensions and improved interpretability.
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).