Automatic Depression Assessment using Machine Learning: A Comprehensive Survey

2026-04-11

AI Summary

Depression assessment suffers from subjectivity, high cost, and clinician shortages, while existing surveys are limited to unimodal behavioral analysis. To address these gaps, this work presents the first systematic survey of multimodal depression-related human behavioral representations—including electroencephalography (EEG), speech, facial expressions, and body gestures—and comprehensively reviews machine learning–based automatic depression assessment (ADA) studies from 2010 to 2023, covering both conventional and deep learning approaches. By synthesizing over 50 publicly available datasets, international benchmark challenges, and state-of-the-art methodologies, we construct the first cross-modal ADA technology map. This map identifies key challenges: modality-specific feature efficacy, model generalization bottlenecks, and dataset biases. Our synthesis establishes an authoritative, structured review framework and delineates concrete research directions for advancing multimodal fusion, model interpretability, and clinical deployment of ADA systems.

Technology Category

Machine Learning: Multimodal LearningComputer Vision: Multi-modal VisionNatural Language Processing: Language Grounding & Multi-modal NLP

Application Category

Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web dataGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Depression is a common mental illness across current human society. Traditional depression assessment relying on inventories and interviews with psychologists frequently suffer from subjective diagnosis results, slow and expensive diagnosis process as well as lack of human resources. Since there is a solid evidence that depression is reflected by various human internal brain activities and external expressive behaviours, early traditional machine learning (ML) and advanced deep learning (DL) models have been widely explored for human behaviour-based automatic depression assessment (ADA) since 2012. However, recent ADA surveys typically only focus on a limited number of human behaviour modalities. Despite being used as a theoretical basis for developing ADA approaches, existing ADA surveys lack a comprehensive review and summary of multi-modal depression-related human behaviours. To bridge this gap, this paper specifically summarises depression-related human behaviours across a range of modalities (e.g. the human brain, verbal language and non-verbal audio/facial/body behaviours). We focus on conducting an up-to-date and comprehensive survey of ML-based ADA approaches for learning depression cues from these behaviours as well as discussing and comparing their distinctive features and limitations. In addition, we also review existing ADA competitions and datasets, identify and discuss the main challenges and opportunities to provide further research directions for future ADA researchers.

Problem

• Subjective and slow traditional depression diagnosis methods
• Lack of comprehensive multi-modal behavior analysis in ADA
• Need for updated ML-based ADA approaches and datasets

Innovation

• Machine learning for automatic depression assessment
• Multi-modal human behavior analysis
• Comprehensive survey of depression-related behaviors