Mar 4, 2024
Services
OS Design
Links
OS Design
GlucoCare AI is a comprehensive prediabetes management application designed with the user at its core. By integrating advanced AI technology and user-centered design principles, we aimed to create an app that not only addresses the growing public health challenge of prediabetes but also empowers users to take control of their health in a personalized and engaging way.
Introduction
Prediabetes poses a significant health challenge. GlucoCare AI is a user-centered mobile app designed to improve prediabetes management by integrating customized health tools within an engaging platform.
Prediabetes is often underdiagnosed and undertreated, leading to significant long-term health risks. Existing digital health solutions lack personalization and user engagement. GlucoCare AI addresses these gaps by providing a highly personalized, engaging, and intuitive tool for active health monitoring and management.
Why We Undertook This Project?
The rise in diabetes and prediabetes cases over the past decades has led to a significant increase in healthcare costs and a decline in quality of life for millions of people. In 2021, approximately 97.6 million Americans were diagnosed with prediabetes. Our goal was to create an application that could help manage this condition more effectively by focusing on lifestyle changes and personalized care, which have been shown to reduce the risk of developing type 2 diabetes by 58%.
Study and Discoveries
Our research phase involved two key components: a comprehensive literature review and in-depth user interviews.
Literature Review
We reviewed twelve research papers on prediabetes and diabetes management using AI. This provided insights into the effectiveness of lifestyle changes and the transformative potential of AI in health management, emphasizing the importance of proactive and personalized care.
User Research
We conducted interviews with three prediabetes patients to understand their experiences and pain points. Key issues included a lack of personalized recommendations and the cumbersome process of logging health data.
Competitive Analysis
We analyzed existing health management apps to identify gaps that GlucoCare AI could fill. Our analysis included popular apps like OneDrop, MyFitnessPal, and MySugr.
Key Findings
Fitness apps do not consider the dietary restrictions required for managing prediabetes.
Manual logging of food details is inconvenient and time-consuming for users.
Most apps focus on daily glucose monitoring, which is not necessary for prediabetes.
Following is the representation of the analysis by rating:
Defining our Objectives
Based on our research and insights, we defined four primary objectives for GlucoCare AI:
Personalize Health Management
Tailor the app to individual users’ health goals and preferences to provide a unique health management plan for each user.
Enable Efficient Meal Tracking
Integrate a meal-scanning feature for quick and accurate dietary monitoring, simplifying the process for users.
Provide Customized Meal Suggestions
Use AI to offer personalized meal options based on users' dietary preferences and nutritional needs.
Support Sustainable Lifestyle Changes
Ensure recommendations promote long-term health benefits by focusing on sustainable lifestyle modifications.
The Design Process
Our design process was iterative, beginning with low-fidelity sketches and evolving into high-fidelity prototypes.
Methodology
We adopted an iterative design process, starting with the conceptualization of user flows and low-fidelity prototyping. This approach allowed us to identify potential usability issues early and refine our designs based on continuous feedback.
Low-Fidelity Designs
We created initial low-fidelity prototypes to visualize the basic layout and functionality, focusing on user flows and key features like the weekly health plan and food tracking.
Design System
To ensure consistency, we developed a comprehensive design system with carefully selected typography, color palette, and adherence to Apple's Human Interface Guidelines.
High-Fidelity Designs
Building on our low-fidelity prototypes, we created detailed high-fidelity prototypes that incorporated our design system, providing a realistic representation of the final product. Key features like meal scanning, health planning, and progress tracking were meticulously designed to enhance user engagement and usability.
User Testing
We conducted comprehensive user testing to gather feedback on the app's usability and functionality.
Testing Methodology
We conducted sessions with three participants, including one pre-diabetic and two non-diabetic users, guiding them through predefined tasks to capture their interactions and reactions.
Feedback Analysis
Positive Feedback
Meal Scanning
Users appreciated the innovative and user-friendly meal scanning feature.
Health Planning
The AI-driven Smart Health Planner received praise for its personalized suggestions.
Progress Tracking
Users found the Visualize Your Progress feature insightful and motivating.
Areas for Improvement
Navigation During Onboarding
Users needed improved navigation, particularly the ability to go back to previous questions.
Nutritional Information Presentation
Users found it challenging to view nutritional information clearly, indicating a need for better segregation of food items.
Iterations and Enhancements
Based on user feedback, we made several enhancements to improve the overall experience.
Changes Implemented:
Increased Contrast in Card Interface
Enhanced visibility and readability, especially under varying lighting conditions.
Editable Meal Components
Added an edit feature to allow users to modify meal details for flexibility and accuracy.
HbA1c Report Upload
Enabled users to upload HbA1c reports, enhancing AI recommendations.
Enhanced Feedback Mechanisms
Integrated a feedback system to gather actionable insights and adjust recommendations accordingly.
AI Transparency
Added explanations for AI suggestions to foster greater user trust.
Learnings and Outcomes
User Insights:
Direct interactions with users provided valuable insights into their needs, driving significant enhancements to the app.
Ethical Considerations:
Transparency in AI decision-making processes was prioritized to build user trust.
Continuous Feedback Loop:
Establishing a continuous feedback loop allowed for iterative improvements, ensuring the app remained responsive to user needs.
Conclusion
The development of GlucoCare AI demonstrated the potential of applying UX design principles to create intuitive and effective health management applications. By involving users throughout the design process and prioritizing transparency and personalization, we created an app that addresses the unique needs of individuals managing prediabetes.
Feedback from user testing was crucial in refining the design and functionality, emphasizing the importance of user involvement. The methodologies and design insights gained from this project provide a valuable framework for future research and development in health technology.
Testimonials
"The GlucoCare AI project team did an exceptional job addressing the needs of individuals with prediabetes. Their use of compelling statistics highlighted the project's urgency, and their thorough review of existing apps and academic papers added depth to their work. The team showcased significant progress from midterm to final presentation, introducing innovative features like AI recommendations and document upload for correcting computer vision errors. Their engaging presentation utilized effective statistics and animations, with each member contributing to a cohesive delivery. Overall, the team effectively conveyed the problems associated with prediabetes and articulated a promising health management solution. Their dedication and effort are truly commendable."
