Harnessing AI in healthcare: Detecting mental health disorders via social media
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The rapid progression of artificial intelligence (AI) is reshaping mental health diagnostics by enabling the analysis of social media data to detect early signs of conditions such as depression, anxiety, and PTSD. This paper presents and evaluates a BERT-based AI framework for identifying mental health disorders. By utilizing advanced natural language processing (NLP) techniques, including BERT for contextual analysis and sentiment analysis tools like VADER, the system identifies linguistic patterns, emotional tone, and behavioral indicators linked to mental health issues. Experimental results reveal the system's strong performance, achieving accuracy rates of 92.68% for identifying depressive episode cases, 93.98% for detecting suicidal ideation, and 87.15% for recognizing anxiety crises, surpassing traditional diagnostic approaches. Additionally, the system demonstrates the capability to forecast mental health emergencies up to 28 days in advance, highlighting its scalability and real-time application. Despite its potential, the study identifies key challenges such as data privacy, ethical considerations, and limitations in handling diverse languages and cultural contexts. To overcome these, the paper advocates for the development of robust privacy safeguards, interdisciplinary collaboration, and advancements in algorithm interpretability and multilingual adaptability. This paper underscores the transformative role of AI in enhancing the mental health diagnostics while emphasizing the importance of ethical and responsible deployment in healthcare systems.
Fields of Study & Subjects
Mental Health via Writing
Sentiment Analysis and Opinion Mining
Emotion and Mood Recognition
Publication details
- Journal / source
- AIP conference proceedings
- Volume
- 3401
- Pages
- 020094–020094
- Publication date
- 1 January 2026
- Language
- EN