To provide an integrated analysis of recent studies on AI-based recognition of facial and micro-expressions for diagnosing mental and neurological disorders, addressing existing knowledge gaps.
Key Findings:
AI technologies show promise in diagnosing mental and neurological disorders through facial expression analysis.
Significant gaps exist in the transition of AI from laboratory research to clinical applications.
Current studies often focus on limited disorders and lack standardized protocols for data collection and evaluation.
Ethical considerations in the application of AI technologies are crucial for their successful implementation.
Interpretation:
The review highlights the potential of AI in enhancing diagnostic accuracy but underscores the need for standardized methodologies and ethical considerations in its application.
Limitations:
Diversity of methods and indicators across studies hindered meta-analysis.
Limited datasets and focus on specific disorders restrict generalizability.
Environmental factors can affect the accuracy of AI algorithms.
Lack of diversity in training datasets limits the applicability of findings.
Conclusion:
The systematic review aims to identify trends, algorithms, datasets, and evaluation methods while addressing limitations and future research directions, particularly emphasizing the need for ethical considerations to improve clinical applicability.
