Issues OCR-ing of Table of content pages

[Ciprian IP.]

There is a major issue with the formatting of lines of after OCR of a page. This issue turns out especially when applying OCR to pages that contain tables and/or Tables of content. The issue manifests as a scrambling between the corresponding lines of what I assume to be a separation into columns (2 in this case). This occurs when selecting the text with the usual selection tool, as in the attached document. 

In other words, when trying to select the text lines, first it picks up a random number of lines, from what seems to be the 1st column grouping, up to a point when it decides to pick up the first line of a second column grouping (see the attached pic). Is there any way out of this strange behaviour?