cant you maybe have any agent like changedetection or huginn use selenium to crawl the page and fetch said info? i think changedetection has rss feeds for changes which include plaintext and/or screenshots.