Who can be an author in a patent from an academic university?

To be an inventor of a patent, you had to have contributed to conception of at least one of the Claims in the patent. Invention under U.S. Patent law generally includes two steps - conception and the act of reducing the invention to practice (note, filing a patent application counts as constructive reduction to practice).

Listed inventors (or authors in your question) must have contributed to the conception part and these means recognizing the ultimate result and developing the means to accomplish that result. Also, it is important to note that you have to focus on the Claim language of the patent, and not the disclosure. Specifically, the disclosure could include your invention, but if it is not in the claims, then you would not be an inventor.

That said, this scenario (where an inventor is left off) is referred to as nonjoinder. The correction depends on whether the patent is a pending patent application or a granted patent. These general processes include filing a lawsuit in Federal court to correct the inventorship.

First, it is good to be clear that ownership in a patent (or patent application) may be different than inventorship. Inventors invent and a presumed to be the owners of the patent absent an agreement assigning their rights, and an owner is the entity that owns the patent (e.g., a university or corporation that was the assigned the patent).

An inventor for a given patent is someone who has conceived of the invention embodied in at least one claim of the patent application (see wikipedia definition).

Inventorship can be corrected after-the-fact at the U.S. Patent And trademark office (see Correcting Inventorship link). However, before correcting inventorship by adding an inventor, it is important to confirm amongst all the listed and unlisted inventors that the added inventor was actually the first to conceive of the claimed invention. Likewise, it is also important to make sure that any added inventor's contribution was included in the claims of the patent and not just in the description. Generally, you will want the cooperation of the patent owner when trying to amend the inventorship of a patent application.

By and large, the answers that precede this one touch the right bases. But there are a few things to say that spin a bit differently than what my predecessors said, so here goes.

First, the legal rules that determine who was a joint inventor of the subject matter of a patent application are different than the principles applied in academia to decide who should be named as the author of a scientific paper. And, to make it even more complicated, the pertinent legal rules under U.S. patent law changed about 25 years ago. Before 1984, someone would not be a joint inventor of a patent unless he contributed conceptually to each of the claims in the patent. Since then, it's enough for a joint inventor to have contributed to at least one of the claims. In the case of Ethicon v. U.S. Surgical (copy attached via a link below), a worker named Choi who had contributed but a minor aspect of the overall invention nevertheless was held to be a joint inventor, even though he was omitted from the patent application. And since he had not transferred his ownership interest to Ethicon, he had the power to license his interest to U.S. Surgical and thereby prevent the latter from being sued for infringement.

That said, the standard of proof for the facts of joint inventorship is a stringent one. Although we argued that my client Richard Levin was in a similar situation to Choi, the Court of Appeals for the Federal Circuit a year ago held otherwise. See the Symantec case attached via a link below.

I would also point out that there isn't uniformity among universities as to whether inventors named on a university-owned patent receive any remuneration if the patent is licensed. Some say no. Others funnel back some money to the department where the work was done. And some universities do channel a portion of licensing revenues to the named inventors. It is worthwhile for someone in the situation you describe to review the school's patent policy, which might be stated in a faculty handbook for example.

You understand, of course, that this discussion of the various principles that might apply is not legal advice and that no attorney-client relationship is created by this communication.

Assuming the omission was a mistake, it can probably be corrected with a fairly straight forward petition to the patent office. I would call the attorney that prosecuted the application and tell him or her that you consider yourself to be a co-inventor.
The University is probably the assignee so it's doubtful that you'll receive licensing compensation if you are added as a co-inventor. You most likely signed an employment agreement that specified that. If that's your goal, you should speak with a local employment attorney.