Deep Learning-Based Facial Expression Recognition for the Elderly: A Systematic Review

2025-02-04

AI Summary

This study addresses critical challenges in elderly facial expression recognition (FER)—including data scarcity, class imbalance, and age-specific expression variations—through a systematic review of 31 deep learning studies published between 2014 and 2024. Employing structured literature analysis, we identify three pervasive bottlenecks: (1) lack of age-inclusive benchmark datasets, (2) insufficient lightweight model adaptation for resource-constrained edge deployment, and (3) absence of explainable AI (XAI) methods. We propose and empirically evaluate a synergistic multimodal fusion framework with privacy-preserving mechanisms, assessing CNNs and their lightweight variants, age-aware data augmentation strategies, XAI techniques, and cross-modal modeling approaches. Results reveal both efficacy and limitations across these dimensions. Finally, we derive three actionable recommendations: (1) constructing high-diversity, age-stratified FER benchmarks; (2) integrating XAI modules to enhance clinical interpretability and trust; and (3) developing edge-optimized lightweight models—thereby providing a clear technical roadmap for deploying FER in real-world elderly affective health monitoring.

Technology Category

Computer Vision: Biometrics, Face, Gesture & PoseCognitive Modeling & Cognitive Systems: Affective ComputingMachine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

User Modeling, Personalization and Recommendation: Fairness-aware retrieval and rankingGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

The rapid aging of the global population has highlighted the need for technologies to support elderly, particularly in healthcare and emotional well-being. Facial expression recognition (FER) systems offer a non-invasive means of monitoring emotional states, with applications in assisted living, mental health support, and personalized care. This study presents a systematic review of deep learning-based FER systems, focusing on their applications for the elderly population. Following a rigorous methodology, we analyzed 31 studies published over the last decade, addressing challenges such as the scarcity of elderly-specific datasets, class imbalances, and the impact of age-related facial expression differences. Our findings show that convolutional neural networks remain dominant in FER, and especially lightweight versions for resource-constrained environments. However, existing datasets often lack diversity in age representation, and real-world deployment remains limited. Additionally, privacy concerns and the need for explainable artificial intelligence emerged as key barriers to adoption. This review underscores the importance of developing age-inclusive datasets, integrating multimodal solutions, and adopting XAI techniques to enhance system usability, reliability, and trustworthiness. We conclude by offering recommendations for future research to bridge the gap between academic progress and real-world implementation in elderly care.

Problem

• Deep learning-based facial expression recognition for elderly.
• Addressing elderly-specific dataset scarcity and class imbalance.
• Enhancing system usability with explainable AI techniques.

Innovation

• Deep learning-based FER systems
• Lightweight convolutional neural networks
• Age-inclusive datasets development