I have tried using Gemini and GPTs to summarize/interpret medical research articles that are often in PDF format. PDF to text conversion that these tools use seem to work well and summaries are on the spot. However, data that is a part of an image (e.g. charts with Events with drug A vs drug B) is lost. Similarly, powerpoint files that present data as images lose most of the information contained in them. Are there tools that can parse graphs/charts from images and convert them into formats (such as a table) that can be used to fine tune LLM?
Thank you!