Background: Theintegration of artificial intelligence (AI) into clinical decision support systems (CDSSs) has significantly enhanced diagnostic precision, risk stratification, and treatment planning. AI models remain a barrier to clinical adoption, emphasizing the critical role of explainable AI (XAI). Methods: This systematic meta-analysis synthesizes findings from 62 peer-reviewed studies published between 2018 and 2025, examining the use of XAI methods within CDSSs across various clinical domains, including radiology, oncology, neurology, and critical care. Model-agnostic techniques such as visualization models like Gradient-weighted Class Activation Mapping (Grad-CAM) and attention mechanisms dominated in imaging and sequential data tasks. Results: However, there are still gaps in user-friendly evaluation, methodological transparency, and ethical issues, as seen by the absence of research that evaluated explanation fidelity, clinician trust, or usability in real-world settings. In order to enable responsible AI implementation in healthcare, our analysis emphasizes the necessity of longitudinal clinical validation, participatory system design, and uniform interpretability measures. Conclusions: This review offers a thorough analysis of the state of XAI practices in CDSSs today, identifies methodological and practical issues, and suggests a path forward for AI solutions that are open, moral, and clinically relevant.

Mental health disorders contribute significantly to the global burden of disease, affecting quality of life and causing disability. These challenges are compounded by inequitable access to timely and effective mental health services, particularly in low-resource settings. Recently, artificial intelligence (AI) has emerged as a transformative tool in mental healthcare, offering novel approaches to enhance diagnosis, personalize treatment, and support continuous patient monitoring. This review explores the current landscape of non-generative AI applications in mental health, focusing on core methodologies such as machine learning, deep learning, and natural language processing. These techniques show promise in improving diagnostic accuracy, enabling adaptive and scalable digital therapy delivery systems, facilitating real-time mental health risk prediction through the analysis of multimodal data. According to our study, the majority of research demonstrated increased therapy personalization and diagnostic accuracy; however, significant challenges still exist due to low dataset diversity, algorithmic bias, and a lack of clinical validation. Ethical considerations and the need for transparent, explainable, and clinician-trustworthy AI are increasingly recognized as critical to successful implementation. Overall, AI-driven methods have strong potential to improve accessibility and effectiveness in mental health treatment, provided future studies prioritize equity, interpretability, and clinical relevance. We ran a narrative review between January 2019 to June 2025, screened in duplicate, and used thematic synthesis across diagnosis, therapy support, and monitoring.