Possible to use OCR + Document Parsing?

[MiguelPilo]

Document Parsing is LIMITED.

Although is an excellent tool to extract data from Files, Its limited only for some file structures
(Invoice-Receipts and W-9).

What if is needed to extract data from different structure files?

Right now Appsheet Monitoring App use Vision AI (Or something similar, I'm not sure) to transform pdf file to an image.png from a "base64 chain".

The question is: Could be saved this image in the main app to use it to train an OCR model and extract other specific data not included in Document Parsing?

Steps could be the nexts:
1. Add document source (Document Parsing)
2. Upload new file to document Parsing
3. Create a bot to get the "image.png" created by Appsheet Monitoring App
4. Save the image in required table and use OCR model previsly created to extract new data.

This way, the limited document analysis function would be released.