DEPRESSION INTENSITY PREDICTION AND PREVENTION VIA SOCIAL MEDIA
DOI: 
SHARE THIS ARTICLE
DOWNLOAD ARTICLE
REVIEW REPORT
CITE THIS ARTICLE
Mental health disorders, particularly depression, constitute a major global health crisis, affecting approximately 280 million individuals worldwide. Conventional diagnostic routes rely on subjective clinical interviews and self-report instruments, which are prone to social desirability bias and accessibility barriers. This paper presents SereneMind — a browser-based, real-time AI framework that continuously monitors a user's emotional and behavioral state through two complementary modalities: (1) a fine-grained facial landmark analysis pipeline built on Google MediaPipe's FaceLandmarker and HandLandmarker models, and (2) a Gemini-AI driven natural language reasoning engine that interprets blendshape metrics in clinical context. The system extracts seven facial action unit proxies — Smile Score, Eye Openness, Blink Detection, Brow Furrow, Inner Brow Raise, Mouth Frown, and Head Orientation — plus Hand Presence detection, aggregating them into a weighted depression intensity score (Low / Moderate / High). The React + Vite + TypeScript front-end streams 30 fps webcam data, maintains a rolling 50-frame history, and renders live sparkline trend charts. Validation across 40 diverse test scenarios yields an 87.5% overall accuracy, with text-grounded AI insights reaching 91.3% contextual reliability. The platform is fully privacy-preserving — all processing occurs client-side with no biometric data leaving the device — making it a clinically responsible, low-barrier screening auxiliary.
Keywords
Article Information
Article Metrics
HOW TO CITE
Copy
Nemade, K. M., Patil, V. B., Bonde, P. Y. & Jadhav, G. J. (2026). Depression Intensity Prediction and Prevention Via Social Media. International Journal of Science, Strategic Management and Technology, 02(04). https://doi.org/10.55041/ijsmt.v2i4.615
Nemade, Khushbu, et al.. "Depression Intensity Prediction and Prevention Via Social Media." International Journal of Science, Strategic Management and Technology, vol. 02, no. 04, 2026, pp. . doi:https://doi.org/10.55041/ijsmt.v2i4.615.
Nemade, Khushbu,Vaishnavi Patil,Pranjal Bonde, and Gayatri Jadhav. "Depression Intensity Prediction and Prevention Via Social Media." International Journal of Science, Strategic Management and Technology 02, no. 04 (2026). https://doi.org/https://doi.org/10.55041/ijsmt.v2i4.615.
References
[2] Patel, V., Chisholm, D., Parikh, R., et al. (2016). Addressing the burden of mental, neurological, and substance use disorders. Lancet, 387(10028), 1672–1685.
[3] Ekman, P., & Friesen, W. V. (1978). Facial Action Coding System: A Technique for the Measurement of Facial Movement. Consulting Psychologists Press.
[4] Girard, J. M., Cohn, J. F., Mahoor, M. H., et al. (2014). Nonverbal Social Withdrawal in Depression: Evidence from Manual and Automatic Analyses. Image and Vision Computing, 32(10), 641–647.
[5] Poria, S., Cambria, E., Bajpai, R., & Hussain, A. (2017). A Review of Affective Computing: From Unimodal Analysis to Multimodal Fusion. Information Fusion, 37, 98–125.
[6] Yang, K., Ji, S., Zhang, T., et al. (2023). Towards Interpretable Deep Learning Models for Knowledge Tracing. Findings of ACL 2023.
[7] Choudhury, S., & Bhatt, C. (2023). LLMs in Clinical Decision Support: Opportunities and Risks. npj Digital Medicine, 6(1), 1–8.
[8] Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. arXiv:1810.04805.
[9] Vaswani, A., Shazeer, N., Parmar, N., et al. (2017). Attention Is All You Need. Advances in Neural Information Processing Systems, 30.
[10] Lugaresi, C., Tang, J., Nash, H., et al. (2019). MediaPipe: A Framework for Building Perception Pipelines. arXiv:1906.08172.
Ethics and Compliance
Indexed In
Similar Articles
Previous Article
