Help :( Backend for Health Data Validation Project
Hey everyone! 👋
I’m working on a project for my studies and could really use some guidance! Here’s what I’m trying to achieve:
1. Objective: Users should upload a picture from their phone (of a document, insurance card, etc.), which would then be analyzed.
2. Tech Stack:
- Frontend: Done in FlutterFlow (design ready).
- Backend: Working on it with Buildship.
3. Functionality:
- The app should use Google Vision or GPT-4 Vision (either is fine) to detect and extract two fields from the uploaded image:
- Health Insurance Name
- IDC-10 Code (a 10-digit code)
- Then, it should check if both values exist in a Google Sheet.
- If both values are found in the sheet, the app should show a “SUCCESS” message in FlutterFlow; otherwise, it should display “NO SUCCESS.”
4. Problem: I’ve been through a lot of resources online, but I still can’t get it to work as needed. I feel stuck, especially in linking Google Vision/GPT-4 Vision to extract specific fields and validate against the Google Sheet.
Has anyone tackled something similar, or does anyone have advice on implementing this backend logic with Buildship? Any tips, tutorials, or code snippets would be a huge help. 🙏
Thanks a ton in advance! 😊
2 Replies
AI Support Bot Information
<@231130069613936650> you can react on the relevant answer (message) with a ✅ in this thread when you think it has been solved, by the bot or by a human!
Anyone can react on this message with a ❌ if the GPT bot is unhelpful or hallucinating answers.
Please note: Team members will review and answer the questions on best efforts basis.
AI Response
I couldn’t find specific templates or nodes in Buildship related to integrating Google Vision or GPT-4 Vision directly. It seems this might be a unique use case that isn't covered by pre-built components.
However, you could consider the following general steps to achieve your project's objectives:
1. Image Upload and Analysis:
- Use Google Vision API to process and analyze images uploaded by the user. This involves sending the image to Google's servers and receiving back detected text and other information.
- If you're considering GPT-4 Vision, you might have to use OpenAI's API if available, but integration can be complex due to potential API limitations regarding image processing.
2. Extracting Specific Information:
- Once you have the output from the vision service, you can write logic to filter out and find the "Health Insurance Name" and "IDC-10 Code" from the recognized text data.
3. Validation Against Google Sheets:
- Use the Google Sheets API to access the spreadsheet and check if the extracted values exist. You can use a REST API or a third-party service to facilitate this integration.
4. Displaying Results:
- Based on the validation results, you can update the UI in FlutterFlow to show either “SUCCESS” or “NO SUCCESS”.
To proceed with the implementation, you might need to write Node.js or Python scripts that leverage the aforementioned APIs, which can then be called from your backend logic in Buildship. If you need deeper technical guidance or code examples, reaching out through Buildship support or online community platforms might be beneficial.
BuildShip AI Support