These have to legally be made available by each hospital for all services they provide (since January 1, 2019).
At the moment they're hard to find, often buried on the hospital's website and not easy comparable to other hospitals in the area.
A kayak-like site for health procedures would be very useful.
So I'm thinking another idea is making a comparison or listing site generator. These sites are very light in terms of tech, heavy in research and operations work, so they're too hard for both the techies and the experts in a field.
re: comparison site generator - I could see that being useful.
I've toyed around with using Google Sheets as the database and the statically regenerating the site every X minutes. Non-techies enter new info via Google forms or directly edit the spreadsheet.
All filter/search is done on the front-end.
I think if you can collect information from patients about the final amount they paid for a given procedure (plus the reference codes) and then use that to find patterns in how those bills correspond to the chargemaster you might be able to create a more useful tool. This becomes much more difficult than just scraping though.
The Trump administration started the process on a bill to force insurance providers to make the details of their contracts transparent earlier this year. If that actually ends up getting done (maybe sometime next year), that would likely be a much more useful source of info to aggregate. Insurance providers will probably try and hide the info as many hospitals did, ready for scrapers to step in.
What you can do is, read the description of each project and do it yourself. Don't look at the solution in the book, then when you are done, compare. If some project excites you further, you can probably extend it so if the book offers a CLI solution, may be write a GUI for it and so on.
The trick with side projects is to build momentum and I hope this technique could be helpful. Have fun doing your side project.
[0] https://github.com/john-bokma/tumblelog
FWIW; I am looking for beta testers ;-)