On the Validity of Head Motion Patterns as Generalisable Depression Biomarkers

2025-05-29

AI Summary

This study investigates whether kinemes—atomic units of head motion—can serve as culturally and environmentally generalizable biomarkers for depression. Method: We systematically evaluate kineme-based representations across three heterogeneous depression datasets from Germany, Australia, and the United States, employing robust kineme modeling to extract discriminative head-motion features, coupled with classical machine learning classifiers (SVM, Random Forest) and rigorous evaluation via k-fold cross-validation and strict cross-dataset transfer testing. Contribution/Results: Kineme features significantly outperform raw head-motion statistics and other visual behavioral cues, achieving the second-lowest mean absolute error (MAE) for depression severity regression on the AVEC2013 benchmark. They demonstrate superior cross-domain stability and clinical applicability in both classification and regression tasks, establishing a generalizable, contactless paradigm for depression assessment grounded in interpretable, biologically plausible motion primitives.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Kernel Methods
• Cognitive Modeling & Cognitive Systems: Affective Computing

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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).

Problem

• Investigates head motion patterns as depression biomarkers
• Assesses model generalisability across diverse cultural datasets
• Compares kineme-based features with other behavioral cues

Innovation

• Uses kinemes as head motion biomarkers
• Validates models across multiple datasets
• Achieves competitive depression severity estimation