VanaAI

Inspiration

A friend working in nursing and assisted living homes shared some of her biggest challenges and pain points throughout the day, inspiring us to develop a solution.

What it does

VanaAI streamlines the process for nurses by offering a voice memo feature that simplifies logging patient interactions. This tool significantly reduces the hours spent on administrative tasks, such as updating Electronic Health Records (EHR), allowing nurses to focus more on patient care.

How we built it

We developed VanaAI using React.js for the frontend and Express/Node for the server-side. The platform is hosted on Firestore NoSQL. Leveraging GPT-4o and Webkit speech recognition, we transcribe voice inputs into concise bullet points, making them compatible with EHR software. The data is then stored in an Excel sheet for monitoring and review before being uploaded to the EHR system.

Challenges we ran into

Initially, we aimed to implement advanced features that could optimize patient placement in nursing home wards based on the platform’s data. However, we realized that nurses might be hesitant to adopt such features right away. So, we decided to start with a minimum viable product that directly addresses their immediate needs.

Accomplishments that we're proud of

We’re especially proud of the accuracy of our AI tools in converting spoken interactions into concise, EHR-compatible bullet points, streamlining the documentation process.

What we learned

We gained valuable insights into the challenges of EHR reporting in nursing homes, particularly how medical records can be flawed due to the overwhelming workload nurses face. This experience underscored the importance of creating tools that truly reduce their burden.

What's next for VanaAI

Our next steps involve engaging with industry professionals—nurses, administrators, and others—to further explore the practical applications and use cases of VanaAI, ensuring it meets the needs of those in the field.

Built With

• auth0
• express.js
• firebase
• gpt-4o
• javascript
• node.js
• nosql
• openai
• react
• typescript
• webkit