PDFBox: Is there a way to extract position of text based on some type of start and end markers?

Question

I have an automated process that produces PDF from HTML+CSS. In my HTML I am marking up the document with highlights for text that I want to later attach annotations to. I am lo