MentalHealth Care Hexabot

Inspiration

Our initial inspiration was to focus on the MentalHealthcare field for its social impact. While using the List block, we discovered it displays all records from the knowledge base table. This led us to create a custom block that filters records based on specific conditions. To optimize the workflow, we integrated a third plugin, enabling users to connect to a doctor's calendar using calendly and book appointments during available time slots.

How we built it

Our workflow begins with the Ollama model, specifically tailored for Mental Health. This model helps users by providing tips and guidance to book appointments and identify their symptoms. It then utilizes a table of doctors stored in the knowledge base. The plugin generates responses based on the data in this table, and the AI model can suggest a doctor. Through the plugin, users can either schedule an appointment with a doctor they already know or find a new one using the doctor plugin.

Challenges we ran into

This was our first experience creating a pipeline that integrates a model with a flow which required time to learn how to switch seamlessly between them and ensure smooth functionality. Additionally, adapting to the new interface presented its own challenges, as it involved interacting with and becoming accustomed to numerous powerful features. While these features are designed to be user-friendly, it still took time to fully understand and utilize them effectively. Finally, debugging during the development of the doctor plugin proved to be a significant challenge. Identifying and resolving errors, checking the source code, and ensuring successful execution required considerable effort and persistence.

Accomplishments that we're proud of

We successfully built a powerful pipeline in a short time, leveraging the tool's low-code approach to simplify traditionally complex tasks, saving time and enhancing efficiency. We integrated advanced features, such as dynamic plugins and seamless AI-model interactions, improving functionality and user experience. We quickly adapted to the tool’s interface, mastering its powerful features to deliver a robust solution. Additionally, our team’s collaboration and problem-solving skills were instrumental in debugging the doctor plugin and ensuring smooth integration. This project not only showcases our technical achievements but also its potential impact on improving mental health care accessibility.

What we learned

We learned how to use Hexabot and develop new plugins related to MongoDB. This includes extracting variables from MongoDB for plugin development using contextType and context, as well as generating various outputs such as lists or buttons. Additionally, we learned how to design workflows with AI models to make the pipeline more efficient and smooth, thanks to this tool.

What's next for Health Care Hexabot

We can integrate advanced features to enhance user guidance and improve their mental health care experience. By incorporating plugins, we enable doctors to maintain a database of patients they have seen, along with detailed information and symptoms extracted from conversations with the AI model. This allows therapists to gain an initial diagnostic based on structured data, streamlining the diagnostic process. Additionally, we can automate visitation processes to improve efficiency and convenience. The system can also suggest the nearest mental health professional using a map API, helping users connect with new doctors based on their location. This comprehensive approach ensures personalized, accessible, and effective mental health support.

Built With

• node.js
• ollama