Inspiration The inspiration for Cognify came from a desire to support dementia patients in maintaining cognitive function and independence. We wanted to create an app that not only engages patients through mind games and reminders but also provides caregivers and doctors with actionable insights to track patient progress. Knowing the impact such a tool could have on the well-being of patients and their families motivated us to bring Cognify to life.
What it does Cognify provides dementia patients with cognitive games and daily reminders that aid memory and routine adherence. Caregivers and doctors can monitor patients’ progress through a dashboard that tracks improvements over time, offering a clear view of cognitive changes between medical visits. The app also includes an image sentiment analysis feature, where patients describe images, and a server evaluates the accuracy of their description to gauge memory recall and cognitive association.
How we built it We developed Cognify using React.js for the frontend, allowing for a dynamic, responsive interface. The backend was built with Django, providing a secure and reliable framework to manage user data and application logic. We used a separate server powered by a Sentence Transformer model for image sentiment analysis, which processes the patient’s description of an image and returns a percentage accuracy score. Chart.js was integrated for visualizing patient progress in each game, helping caregivers and doctors easily interpret changes in cognitive function. We also utilized MySQL with Django to securely store user information and progress data.
Challenges we ran into Integrating the image sentiment analysis server with the main application was a complex task, as we had to ensure consistent communication between the separate servers. Setting up real-time, interactive charts with Chart.js required careful planning to make the data visually intuitive. Another challenge was designing a user-friendly interface for dementia patients while keeping the caregiver and doctor dashboards informative and easy to navigate.
Accomplishments that we're proud of We’re proud of successfully implementing a tool that leverages machine learning to assess image sentiment, offering a unique measure of patient cognitive association. The patient progress tracking with Chart.js also turned out to be a valuable feature, giving caregivers clear and actionable insights into patients' cognitive improvements. Achieving a balanced design suitable for both patients and caregivers was also a significant accomplishment.
What we learned Throughout the development of Cognify, we deepened our understanding of integrating machine learning models with web applications and learned valuable techniques for managing complex data visualizations. Working with React, Django, and MySQL taught us a lot about handling real-time data and designing scalable architectures for healthcare applications. We also learned a lot about dementia in our research and learned about the many ways dementia patients can be helped.
What's next for Cognify Moving forward, we aim to expand Cognify by adding more cognitive games and refining the sentiment analysis to assess various aspects of memory and language. We also plan to implement more advanced progress-tracking metrics and explore real-time data sharing capabilities to give caregivers and doctors instant insights. Our vision is for Cognify to evolve into a comprehensive cognitive support platform for dementia patients and their caregivers.
Log in or sign up for Devpost to join the conversation.