The AI MoodRing project addresses the critical challenge of identifying and supporting students, particularly those with Special Educational Needs (SEN), experiencing emotional distress, where timely intervention is often hindered. AI MoodRing offers a proactive solution leveraging wearable technology and artificial intelligence for early detection and support.

Key Innovations & Design Concepts: The system utilizes a smartwatch with sensors monitoring physiological indicators like Heart Rate Variability (HRV) and Skin Conductance Response (SCR), scientifically linked to stress and emotional arousal. This real-time data, analyzed by a Cloud AI engine against personalized baselines, enables early identification of significant emotional changes. The design integrates sensing, analysis, alerting, guidance, immediate student comfort features (e.g., music, breathing exercises initiated via the device/app), and pathways to professional support.

Technical Principles: Wearable sensors capture HRV and SCR data. This is transmitted wirelessly (Bluetooth/App) to a Cloud platform. The AI engine processes this data, identifies anomalies indicating potential distress based on individual thresholds, and triggers configurable alerts. Notifications with actionable guidance are sent to teachers/parents via a mobile app, which also allows real-time monitoring.

Potential Impact: AI MoodRing aims to significantly improve student well-being by enabling timely intervention, reducing emotional outbursts, and fostering self-regulation skills. It empowers educators with data-driven insights and practical tools, strengthens home-school collaboration through enhanced communication and understanding, and contributes to safer, more supportive learning environments. This system has the potential to establish a new standard for preventative mental and emotional health support in schools.

Our inspiration for AI MoodRing originated from observing the significant emotional challenges faced by students, particularly those with SEN, highlighted by concerning statistics (e.g., rising youth suicide rates in Hong Kong, high parental reporting of SEN child emotional issues). We identified a critical gap within the education system: teachers, often overburdened, lack tools for continuous, real-time observation of individual student emotional states. This leads to delayed detection of distress, missed opportunities for early intervention, and potential escalation into negative incidents. Believing technology could offer a proactive solution, we were motivated to design a system using accessible wearables and AI to provide timely, data-informed support.

Our underlying hypothesis is: By continuously monitoring key physiological indicators (HRV and SCR) linked to emotional states via wearable sensors and analyzing this data with AI, we can reliably detect early signs of significant emotional distress in students, thereby enabling prompt, targeted interventions that improve well-being and mitigate negative outcomes.  

We believe this will succeed because:

Scientific Validity:
The chosen indicators (HRV, SCR) are scientifically recognized markers reflecting autonomic nervous system activity, which is closely tied to emotional arousal and stress.  
Technological Maturity:
Wearable sensor technology and AI-driven pattern analysis are sufficiently advanced to implement this monitoring and detection effectively.
Directly Addresses Need:
The system targets the crucial bottleneck of delayed identification by providing objective, real-time alerts, overcoming limitations of human observation alone in busy classrooms.
Holistic Approach:
Success hinges not just on detection but on the integrated response – providing immediate comfort to students, actionable guidance to teachers, and clear pathways to further support, creating a practical and relevant support mechanism within the educational context.

We will leverage smartwatch wearable sensors (PPG for Heart Rate Variability - HRV, EDA for Skin Conductance Response - SCR), Bluetooth Low Energy for data transmission to dedicated mobile apps (iOS/Android), and a secure Cloud platform (e.g., AWS/GCP) hosting our AI/Machine Learning engine for real-time physiological data analysis. Push notifications will deliver alerts. Required resources include funding (R&D, hardware, cloud services), skilled personnel (engineers, AI specialists, UX designers, educational psychologists), partnerships with schools for pilot testing, and potentially ethical access to relevant datasets for initial AI model training. We plan to validate market demand through surveys and focus groups with teachers, parents, and school administrators, analysis of EdTech adoption trends, and crucially, assessing engagement and perceived value during pilot programs.

The core functionalities are:

1. Continuous HRV/SCR monitoring.
2. AI-driven analysis for detecting emotional distress anomalies against personalized baselines.
3. Configurable real-time alerts and notifications for teachers and parents via app.
4. Dashboards for authorized monitoring.
5. Actionable guidance prompts for teachers upon alert.
6. Integrated immediate comfort stimuli for students (e.g., app-guided breathing/music).
7. Secure data logging and reporting.
8. Facilitation of professional referral pathways.

To ensure a positive user experience, we will focus on intuitive, user-friendly interfaces tailored to each role (especially simple/non-intrusive for students), comfortable device wearability, transparent communication regarding data use and privacy, and accessible support. Performance metrics will include: AI alert accuracy (precision/recall validated against teacher logs/self-reports), system reliability (uptime, battery life), user engagement rates (app usage), user satisfaction surveys, and qualitative feedback on perceived effectiveness in reducing distress incidents during pilots.

We acknowledge question 2b regarding Technical Implementation and Performance. Based on the provided guidelines, this section is designated for projects in Streams 3 and 4. As our project, AI MoodRing, is classified under Stream 2, we understand this specific question does not apply to our submission.

AI MoodRing represents an innovative and creative solution by shifting the approach to student emotional support from primarily reactive observation to proactive, data-driven early detection. Traditional methods often identify distress only after it manifests behaviourally, leading to delayed interventions. Our core innovation is the application of continuous physiological monitoring (HRV, SCR) combined with AI analysis specifically within the educational context to anticipate and identify potential emotional dysregulation before it escalates. This preventative focus, leveraging accessible technology for real-time insight, creatively addresses the limitations of current support systems.

The project demonstrates innovation and creativity through several key aspects:

1. Integrated Ecosystem:
It uniquely combines wearable sensing, AI-powered analytics, multi-stakeholder communication (alerts to teachers/parents), actionable guidance for educators, and immediate, automated comfort interventions delivered directly to the student (e.g., calming stimuli via the device/app). This holistic integration from detection to initial support is a novel design.

2. Personalized Sensitivity:
Utilizing AI to establish individual physiological baselines and dynamic alert thresholds moves beyond generic indicators, offering a more personalized and potentially more accurate detection system.

3. Direct Student Empowerment:
Creatively using the technology not just to monitor, but to provide students with immediate, accessible tools (like guided breathing or music) for in-the-moment self-regulation, fostering agency.

These innovative elements directly enhance effectiveness in addressing user challenges. The real-time detection and alerts combat the problem of delayed intervention. The integrated guidance supports teachers by reducing response uncertainty and burden. The immediate comfort features directly address student distress and promote coping skills. By combining these features, AI MoodRing offers a more timely, targeted, and multi-faceted approach to improving student emotional well-being in schools.

Our scalability strategy leverages a cloud-native architecture (e.g., AWS, GCP) utilizing auto-scaling resources, managed databases, and load balancing to dynamically handle increasing numbers of users and data throughput. Potential bottlenecks like AI processing demand during peak school hours or mass notification delivery will be mitigated through optimized algorithms, efficient database indexing, asynchronous processing, and potentially serverless functions for elastic compute power. A modular system design allows individual components (e.g., data ingestion, AI analysis, notification service) to be scaled independently. Operationally, we plan scalable onboarding through online resources and potentially train-the-trainer programs for schools.  

Environmental sustainability is considered by prioritizing energy-efficient wearable devices and cloud providers committed to renewable energy. We aim for durable hardware designs and will explore partnerships for responsible end-of-life device recycling or refurbishment programs. Software efficiency will be optimized to minimize computational load and energy consumption.

Long-term user engagement will be fostered by continuously demonstrating value – providing reliable alerts, useful guidance, and effective comfort features. We will implement strong user feedback mechanisms (surveys, in-app feedback) to drive regular feature updates and improvements, enhancing personalization and usability. Building a knowledge base and responsive support system will also be key.

Adaptability to evolving user needs is ensured through agile development methodologies, allowing iterative improvements based on ongoing feedback and pilot program data. The modular architecture facilitates integration of new sensor types or evidence-based therapeutic modules as research advances. Continuous monitoring of user requirements and educational best practices will guide future development, ensuring the solution remains relevant and effective.

AI MoodRing directly addresses pressing social issues, including the youth mental health crisis and educational equity gaps faced by SEN students. By facilitating early, objective detection of emotional distress, our solution aims to reduce stigma around mental health challenges and prevent negative incidents leading to student exclusion.

This project enhances the lives of its primary beneficiaries:

- Students: Receive timely support, gain tools for self-regulation, potentially experience reduced anxiety and fewer crises, fostering better educational engagement.

- Teachers: Benefit from reduced stress from managing crises, gain insights for effective interventions, and feel more confident supporting diverse learners.

- Parents: Achieve deeper understanding of their child, improve home-school partnership, and receive guidance towards resources.

Our solution aligns with broader social goals of equity and inclusion by providing personalized, data-informed support tailored to SEN students' needs, promoting equitable success. Early intervention can prevent issue escalation leading to marginalization.

To measure social impact, we will use metrics such as:

- Reduction in school-reported critical incidents or disciplinary actions among users.
- Pre/post-intervention changes in validated well-being or anxiety survey scores (where feasible).
- Increased teacher and parent self-reported confidence in supporting students.
- Qualitative data from case studies and testimonials.
- System adoption and retention rates.

Responsiveness to evolving community needs will be maintained through continuous dialogue. We plan advisory groups with diverse stakeholders, regular feedback sessions, and iterative adaptation of system features and protocols based on community input and impact data.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.