AI-based emotion recognition in dementia through facial expression: A scoping review

Area of Science:

• Gerontology
• Artificial Intelligence
• Psychiatry

Background:

• Emotion assessment is crucial for dementia patient well-being and care planning.
• Traditional methods for emotion assessment in dementia are subjective and time-consuming.
• AI-based facial expression analysis offers a potential objective and efficient alternative.

Purpose of the Study:

• To conduct a scoping review on the use of AI-based facial expression analysis for emotion recognition in dementia patients.
• To identify current applications, methodologies, and challenges of AI in dementia emotion assessment.

Main Methods:

• A scoping review using the SPIDER strategy was performed.
• Searches were conducted across five major databases (PubMed, Scopus, PsycInfo, ProQuest, IEEE Xplore) with snowballing.
• Data from 11 selected studies were extracted and analyzed, focusing on AI tools and clinical outcomes.

Main Results:

• Eleven studies were included, primarily utilizing deep neural networks for emotion recognition.
• AI applications spanned early detection, diagnosis, and intervention assessment.
• Most studies used pre-existing datasets, highlighting a need for original data collection.

Conclusions:

• AI tools demonstrate significant potential to enhance emotion recognition and improve dementia care.
• Standardized protocols for data collection and processing are essential for advancing AI in this field.
• Addressing dataset limitations, prioritizing ethical considerations, and fostering interdisciplinary collaboration are critical for future development.