MindEase: AI-Powered Digital Mental Health and Psychological Support System for Students

Abstract

With the growing concern over mental health challenges among students, the need for accessible psychological support systems has become critical. Traditional counselling centres often face resource shortages, leaving many students without timely assistance. This study presents MindEase, an AI-powered digital mental health platform designed for university students. The system follows a three-tiered hybrid model that integrates automated and human-driven interventions. The first tier features an AI chatbot developed using Rasa, capable of conducting validated assessments such as PHQ-9 and GAD-7 to evaluate stress, anxiety, and depression levels. Additional components include an anonymous peer discussion forum and a secure Institutional Analytics Dashboard for administrators. Built with React, React Native, Node.js, and MongoDB, the platform ensures privacy through JWT authentication and AES-256 encryption. Experimental deployment highlights its scalability, accessibility, and potential to enhance mental health support while reducing the burden on institutional counseling services.

Introduction

The modern higher education environment, shaped by academic pressure, socio-economic changes, and post-pandemic hybrid learning, has exacerbated mental health challenges among students in India. Mental health issues are widespread, costly, and under-resourced, with a severe shortage of professionals and systemic barriers to care.

MindEase is a mobile platform designed to address these gaps through a hybrid human-AI support system using a stepped-care model. It combines:

• Tier 1: AI-driven psychoeducation, mood tracking, and self-help tools.

• Tier 2: Screening, peer support, and guided self-help modules.

• Tier 3: Professional counselor intervention for moderate to high-risk cases.

The platform leverages conversational AI, peer forums, and an institutional analytics dashboard to provide rapid, scalable, and secure mental health support while enhancing human empathy and proactively identifying at-risk students.

Technically, MindEase employs a modular service-oriented architecture with React, Node.js, Rasa NLU/NLG, MongoDB, TensorFlow, and strong encryption standards (AES-256, JWT) to ensure performance, scalability, and regulatory compliance. NLP models handle intent recognition, sentiment analysis, and risk detection to trigger appropriate interventions.

MindEase bridges gaps in existing digital platforms by integrating clinical workflows, prioritizing security, and offering proactive campus-wide mental health monitoring.

Conclusion

A safe, scalable, and clinically based digital mental health platform designed especially for college students\' needs is successfully implemented by MindEase. Its main technical contributions are the combination of a clinically responsible stepped-care model with robust conversational AI screening (using PHQ-9 and GAD-7). Strict data security measures, such as AES-256 encryption and JWT authentication, strengthen this infrastructure, which is also carefully paired with institutional-level data analytics for proactive intervention.

References

[1] NFHS-5 Mental Health Data and the World Health Organization (WHO). (Combined information on economic impact, treatment gap, and prevalence). [2] PMC, 2023, \"An Overview of Chatbot-Based Mobile Mental Health Apps\" [3] \"Human–AI Collaboration Enables More Empathic Conversations in Text-Based Peer-to-Peer Mental Health Support,\" arXiv, 2022, by A. Sharma et al. [4] WICHE, \"E-Mental Health Solutions in Higher Education,\" CPEMH, 2024. [5] \"Peer Support for Mental Illness in India,\" Epidemiol Psychiatr Sci, PMC, 2018. [6] Attract Group, \"Mental Health App Development Guide.\" [7] S. Sarkar, Kaggle, Sentiment Analysis for Mental Health Dataset. [8] Bielen, T. et al., \"Chatbots for Mental Health Screenings.\" [9] M. K. Tavakoli and associates, \"Comparison of the models\' validation accuracy and F1 score.\" [10] \"Evaluation of the Validity and Potential Utility of the ChatGPT-4 in Measuring Anxiety and Depression\" by H. S. Lee and associates. PMC, 2024. [11] M. A. W. Lin et al., \"Evaluation of an AI Chatbot for Depression Screening,\" PMC, 2023. [12] Moldstud, \"Improving User Authentication in Healthcare Apps through Encryption\".

Copyright

Copyright © 2025 Renu Chaudhary, Rachna , Diya Gupta, Manav Garg, Khushi Bansal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.