Pre-Grant Publication Number: 20080077604
Collaborate on the process of community review for this application. Posting will not be forwarded to the USPTO. Flagging a post as an ACTION ITEM signals further research. Flagging SPAM and ABUSE helps to manage discussion. Placing double brackets around a reference to a claim or prior art will create a hyperlink to the original ex. [[claim 1]] and [[prior art 2]].

Please review the Community Code of Conduct prior to posting

Discussion (7)
  Facilitator's Comment     Action Item
  Show without Noise
6
Steven Pearson (2 months ago)
"Building an anonymized catalogued radiology museum in PACS: a feasibility study" also looks potentially interesting as prior art, but I cannot find a free copy of the entire treatise, only this abstract: http://bjr.birjournals.org/cgi/content/abstract/79/944/666
which by itself doesn't appear to offer more than the existing examples of prior art here.
5
Steven Pearson (2 months ago)
The application states that "the primary limitation in the prior art methods is the inability to provide a masking for the secondary object data that may comprise identification details of the medical object." Presumably its contribution may be centered in this area. I notice that the description does suggest that one can anonymize various kinds of secondary data, but I'm not satisfied it does a good job teaching how to do that. I found herein two approaches for non-text data: (1) convert to text, anonymize, convert back, and (2) define what a confidential wave-form looks like, and find/replace it in the wave-form data. I can understand how (1) works, at least where the secondary data type already has conversions to and from text defined (as I don't see any such method elaborated here). It may just be my ignorance of this specfic method, but the description for (2) in the application is not very satisfying. E.g. the application says "The object data being present in the frequency domain waveform format, the element of the object data matching a predetermined frequency can be identified as a confidential identification data." I guess confidential data somehow has a unique frequency? Anyway, aside from these methods, the application seems to simply go back again and again to the notion of "find confidential data and replace it", which does not seem by itself to be a novel concept.
Steven Pearson (2 months ago)
For what it's worth, I should have also said that in the case of the wave form, the method also involves converting "from a time domain waveform format to a frequency domain waveform format" and back.
4
Susan Murray (2 months ago)
Claim 1 could be read as being broad enough to cover manual redaction of hard copy documents. In that regard, it's reasonable to raise HIPAA itself. However, note that the applicant has narrower depedent claims that might warrant some attention as well as more detailed disclosure which may be able to support more narrowly drafted claims, particularly relating to specifics about data structures and filtering mechanics. Some focus in that direction might be useful.
3
Steven Pearson (3 months ago)
The submittor of Prior Art Reference 202 argues that this application essentially repeats the requirements of the HIPAA law. Attempting to patent the common implementation of a standard is not without precedent :-) However, as the law's been in effect for several years already, it seems that there should be prior art out there that would defeat any trivial implementation of it.
2
Steven Pearson (3 months ago)
An internet search on "DICOM anonymize" led to a bunch of hits for tools that apparently anonymize DICOM data.
1
Steven Pearson (3 months ago)
Ok, so this is essentially about "anonymizing" medical records. The terms "object" or "medical object" refer to a person/patient. The gist of the invention is to locate and replace confidential identification data in a medical document.