I had the following study on record in PacsOne:
PatientID=Simon; PatientName=Wright
I send out a new study with:
PatientID=Simon; PatientName=Lehrman
That last study has been added in the db under Simon/Wright. Not good!
We need to fix that...
I guess in human medicine (I'm working for a veterinary practice), the PatientID is some numeric string and more likely to be unique.
But in my case PatientID is the animal's name. So guess how many "Felix" the cat I have... Pet owners really don't have that much imagination.
Does PatientID has to be unique???
Yes, patient ID MUST be unique and that's one of the reasons most facilites use HIS/RIS which is responsible for generating unique patient IDs within the network.
Another advantage of using unique patient IDs would be to allow multiple patients with identical names to co-exit in the database. As you mentioned, it's quite common to have multiple patients with the same name, but their patient IDs can be globaly unique. (There're many unique id generation schemes that will guarantee the uniqueness of the resulting id strings.)
Another advantage of using unique patient IDs would be to allow multiple patients with identical names to co-exit in the database. As you mentioned, it's quite common to have multiple patients with the same name, but their patient IDs can be globaly unique. (There're many unique id generation schemes that will guarantee the uniqueness of the resulting id strings.)
This scheme does not guarantee the uniqueness either. You can do a Google search on "Globally Unique ID Genereation" and I'm pretty sure there're many such algorithms or even existing programs that generate unique ids.Antw1 wrote:I guess from now on my scheme of uniqueness is going to have "Felix#1", "Felix#2", and so on...
Another simple solution is to use the date and time when a patient is first registered, e.g., YYYY-MM-DD-HH-MM, assuming your facility is not too busy to admit multiple patients at the very same minute.
Strange error message when conflicting patientID/name:
Here's a hint in the jungle of strange errormessages: 
I got a strange errormessage when trying to store a patient with the same patientID but other name: "Error! No space on PACS-node. Inform responsible person of remote system to care about free space."
It turned out to be a patient previously registered with the same patientID, but misspelled name. Did take me a while to understand that, because the study that generated the error looked correct, the problem was the previous study wich was misspelled.
I understand the strange errormessage on my workstation, because what could the PACS say, there aren't so many errormessages to send as response to a workstation trying Dicomstorage with bad data...
But still very confusing and timeconsuming to solve...
I got a strange errormessage when trying to store a patient with the same patientID but other name: "Error! No space on PACS-node. Inform responsible person of remote system to care about free space."
It turned out to be a patient previously registered with the same patientID, but misspelled name. Did take me a while to understand that, because the study that generated the error looked correct, the problem was the previous study wich was misspelled.
I understand the strange errormessage on my workstation, because what could the PACS say, there aren't so many errormessages to send as response to a workstation trying Dicomstorage with bad data...
But still very confusing and timeconsuming to solve...
No don't add private. It's ok as it is. Though it would be usefull with a more easy-to-read error log available from the web interface. Maybe with possibility to use different filters, in order to select the type of errors you want to see. Could be combined with a link to the PacsOne site with a on-line help for common causes for a certain error message. But I dont think that is a priority projekt...
Sometimes I'm missing a direct window for debugging from PacsOne, that is a positive log that also tells me everything good.... Every transaction, like from a Dicom sniffer. "Verbose mode" or whatever you could call it, don't know how to invoke/put it away in a nice manner though. This mode could be good in a lot of situations, because PacsOne is very silent when you are in trouble, its hard to see what happens in the "black box".
Sometimes I'm missing a direct window for debugging from PacsOne, that is a positive log that also tells me everything good.... Every transaction, like from a Dicom sniffer. "Verbose mode" or whatever you could call it, don't know how to invoke/put it away in a nice manner though. This mode could be good in a lot of situations, because PacsOne is very silent when you are in trouble, its hard to see what happens in the "black box".