arXiv:2511.15675v2 Announce Type: replace 
Abstract: Depression is a prevalent global mental health disorder, characterised by persistent low mood and anhedonia. However, it remains underdiagnosed because current diagnostic methods depend heavily on subjective clinical assessments. To enable objective detection, we introduce a gold standard dataset of 103 clinically assessed participants collected through a tripartite data approach which uniquely integrated eye tracking data with audio and video to give a comprehensive representation of depressive symptoms. Eye tracking data quantifies the attentional bias towards negative stimuli that is frequently observed in depressed groups. Audio and video data capture the affective flattening and psychomotor retardation characteristic of depression. Statistical validation confirmed their significant discriminative power in distinguishing depressed from non depressed groups. We address a critical limitation of existing graph-based models that focus on low-frequency information and propose a Multi-Frequency Graph Convolutional Network (MF-GCN). This framework consists of a novel Multi-Frequency Filter Bank Module (MFFBM), which can leverage both low and high frequency signals. Extensive evaluation against traditional machine learning algorithms and deep learning frameworks demonstrates that MF-GCN consistently outperforms baselines. In binary classification, the model achieved a sensitivity of 0.96 and F2 score of 0.94. For the 3 class classification task, the proposed method achieved a sensitivity of 0.79 and specificity of 0.87 and siginificantly suprassed other models. To validate generalizability, the model was also evaluated on the Chinese Multimodal Depression Corpus (CMDC) dataset and achieved a sensitivity of 0.95 and F2 score of 0.96. These results confirm that our trimodal, multi frequency framework effectively captures cross modal interaction for accurate depression detection.

تقدم دراسة جديدة شبكة الالتفاف البيانية متعددة الترددات (MF-GCN) المصممة للكشف عن الاكتئاب ثلاثي الأبعاد من خلال دمج بيانات تتبع العين، والوجه، والسمات الصوتية. يستخدم هذا النهج المبتكر مجموعة بيانات تضم 103 مشاركًا تم تقييمهم سريريًا، بهدف تقديم قياسات موضوعية للأعراض الاكتئابية، التي غالبًا ما يتم تشخيصها بشكل غير كافٍ بسبب الاعتماد على التقييمات الذاتية.

Un nuevo estudio presenta la Red de Convolución Gráfica Multi-Frecuencia (MF-GCN), diseñada para la detección de depresión tri-modal mediante la integración de datos de seguimiento ocular, faciales y acústicos. Este enfoque innovador utiliza un conjunto de datos de 103 participantes evaluados clínicamente, con el objetivo de proporcionar medidas objetivas de los síntomas depresivos, que a menudo son subdiagnosticados debido a la dependencia de evaluaciones subjetivas.

Une nouvelle étude présente le Réseau de Convolution Graphique Multi-Fréquence (MF-GCN), conçu pour la détection de la dépression tri-modale en intégrant des données de suivi oculaire, faciales et acoustiques. Cette approche innovante utilise un ensemble de données de 103 participants évalués cliniquement, visant à fournir des mesures objectives des symptômes dépressifs, souvent sous-diagnostiqués en raison de la dépendance à des évaluations subjectives.

A new study introduces the Multi-Frequency Graph Convolutional Network (MF-GCN), designed for tri-modal depression detection by integrating eye-tracking, facial, and acoustic features. This innovative approach utilizes a dataset of 103 clinically assessed participants, aiming to provide objective measures of depressive symptoms, which are often underdiagnosed due to reliance on subjective assessments.

MF-GCN: A Multi-Frequency Graph Convolutional Network for Tri-Modal Depression Detection Using Eye-Tracking, Facial, and Acoustic Features

arXiv:2512.06447v1 Announce Type: new 
Abstract: Depression poses serious public health risks, including suicide, underscoring the urgency of timely and scalable screening. Multimodal automatic depression detection (ADD) offers a promising solution; however, widely studied audio- and video-based ADD methods lack a unified, generalizable framework for diverse depression recognition scenarios and show limited stability to missing modalities, which are common in real-world data. In this work, we propose a unified framework for Stable Cross-Domain Depression Recognition based on Multimodal Large Language Model (SCD-MLLM). The framework supports the integration and processing of heterogeneous depression-related data collected from varied sources while maintaining stability in the presence of incomplete modality inputs. Specifically, SCD-MLLM introduces two key components: (i) Multi-Source Data Input Adapter (MDIA), which employs masking mechanism and task-specific prompts to transform heterogeneous depression-related inputs into uniform token sequences, addressing inconsistency across diverse data sources; (ii) Modality-Aware Adaptive Fusion Module (MAFM), which adaptively integrates audio and visual features via a shared projection mechanism, enhancing resilience under missing modality conditions. e conduct comprehensive experiments under multi-dataset joint training settings on five publicly available and heterogeneous depression datasets from diverse scenarios: CMDC, AVEC2014, DAIC-WOZ, DVlog, and EATD. Across both complete and partial modality settings, SCD-MLLM outperforms state-of-the-art (SOTA) models as well as leading commercial LLMs (Gemini and GPT), demonstrating superior cross-domain generalization, enhanced ability to capture multimodal cues of depression, and strong stability to missing modality cases in real-world applications.

تم اقتراح إطار عمل جديد للتعرف المستقر على الاكتئاب عبر المجالات، يسمى SCD-MLLM، لتحسين الكشف التلقائي عن الاكتئاب من خلال دمج مصادر بيانات متنوعة مع معالجة التحديات التي تطرحها النماذج المفقودة. يهدف هذا الإطار إلى تحسين الاستقرار والدقة في التعرف على الاكتئاب في السيناريوهات الواقعية حيث قد تكون البيانات غير مكتملة.

Se ha propuesto un nuevo marco para el Reconocimiento Estable de la Depresión Inter-Dominios, denominado SCD-MLLM, para mejorar la detección automática de la depresión mediante la integración de diversas fuentes de datos, abordando al mismo tiempo los desafíos que presentan las modalidades faltantes. Este marco busca mejorar la estabilidad y precisión del reconocimiento de la depresión en escenarios del mundo real donde los datos pueden estar incompletos.

Un nouveau cadre pour la reconnaissance stable de la dépression inter-domaines, nommé SCD-MLLM, a été proposé pour améliorer la détection automatique de la dépression en intégrant diverses sources de données tout en abordant les défis posés par les modalités manquantes. Ce cadre vise à améliorer la stabilité et la précision de la reconnaissance de la dépression dans des scénarios réels où les données peuvent être incomplètes.

A new framework for Stable Cross-Domain Depression Recognition, named SCD-MLLM, has been proposed to enhance automatic depression detection by integrating diverse data sources while addressing the challenges posed by missing modalities. This framework aims to improve the stability and accuracy of depression recognition in real-world scenarios where data may be incomplete.

MF-GCN: A Multi-Frequency Graph Convolutional Network for Tri-Modal Depression Detection Using Eye-Tracking, Facial, and Acoustic Features

Was this article worth reading? Share it

LucidQuery AI

DBTAI

ChatOne