BIR: determining if a slice of a sagittal image is to the left or right of the medial

I'm implementing the requirement for BIR that says I should indicate whether the current sagittal image is on the left or right hand side of the medial sagittal line. It's easy enough to determine if an MRI image is sagittal from the directional cosines. Is this simply a matter of checking the Image Position Patient attribute and if it is > 0 the image is to the left and if it is < 0 it is to the right? I'm assuming for simplicity that the patient positions is HFS. Best regards, Bill Prada

radiation dose exploitation: MPPS or SR

Hi all, I'm willing to make an extended use of the Radiation Dose provided by different modalities along the Hospital. Actually, this information is not widely used or not as much as it could be, and due to its potential interest I've decided to study the way how to exploit it. I've few knowledge about radiation dose information and its exchange among PACS environment but all I have seen so far by now is that the current way for communicating this type of info (MPPS messages) is no longer "recommended" anymore but instead to use proper SR objects (at least on XRay and CT scenario

Query modality (Scanner) from other DICOM nodes

Hi All, I would like to know your view on Querying modality from other DICOM nodes. Is this really needed? Should the modality (scanner) support DICOM Q/R? My impression is that, this can lead to a incomplete/inconsistant data in a hospital workflow. For example, if a system retrieves study from modality while the scan is half way, then it provides only an incomplete data. Is there any workflow in which the query/retrieve to a modality needs to be supported in real hospital workflow. Thank you, Sujesh

help with GE MRI image voxel positions

Dear All, The question is regards GE MRI image. I have two MRI images of the same slice. They were acquired with different FOV: one 230mm X 230 mm, and the other 160mmX 160mm. But they are not co-centered. I know the ImagePositionPatient and ImageOrientationPatient for the first MRI image from the DICOM header, and the Center (cx, cy, cz in mm) of the 2nd image from the P-file (GE's own format). I want to find out the position of the Center of the 2nd image on the 1st image. The algorithm is simple: 1. get the IPP2C vector -- a vector from ImagePositionPatient to Center

digital signature and watermark

Hi all, I'm sorry for my bad English but I need to know one thing. In this period I'm developping a software that could sign a DICOM image with a digital signature and then could verify the integrity and the authentication of the image, The standard suggests to hide the digital signature in the tags of the DICOM file, but for my goals this method is not enough secure. So I decide to test the technique of the watermark that seems to be the best for my goals. My question is: talks the standard about this in some part? If yes, defines the standard a limit of the minimum PSNR (peak signal

DX: ViewPosition - how to obtain?

Greetings, I am looking at some DICOM data for DX images. I have a single series listed as PA, which contains 3 images. In one image: SeriesDescription is PA But, it is obvioulsy an Oblique view, and both SeriesDescription and ProtocolDescription DICOM data confirm this with the value of OBL. However, ViewPosition (0018,5101) is not present, so how can I accurately determine if the ViewPosition is RLO or LLO from the DICOM data alone? For this example (0020,0020)PatientOrientation is L\F (0020,0062) ImageLaterality is U but, (0020,0060) Laterality is also not presen

Type of DICOM Attributes

Hi All, I wanted to know the DICOM tag TYPES for the following tags, Please help me [0x00080092] =3D eST; //Referring Physician's Address [0x00080094] =3D eSH; //Referring Physician's Telephone number [0x00100050] =3D eSQ; //Patient's Insurance Plan Code Sequence [0x00101005] =3D ePN; //Patient's Birth Name [0x00101040] =3D eLO; //Patient's Address [0x00101060] =3D ePN; //Patient's Mother's Birth Name [0x00101090] =3D eLO; //Medical Record Locator [0

What is a good use case for accepting 1 of N transfer syntaxes from an A-ASSOCIATE-RQ

An SCU can request more then 1 transfer syntax in a presentation context. For example: JPEG Lossless, Explicit Little Endian and Implicit Little Edian. The SCP must select one of the proposed transfer syntaxes or reject the requested abstract syntax. A question is which algorithm should the SCP use to select which transfer syntax to accept? - A) The SCP selects the first transfer syntax that it can support. The SCU can anticipate on this behavior and sort the requested transfer syntaxes in a certain way. - B) The acceptor has its own preference. It selects for example a lossless

Possible to determine T1 vs. T2 from DICOM tags?

Greetings, I am trying to extract as much data as possible from both DICOM and HL7 (study description, series description, etc.) for all images stored on our PACS, in order to further manipulate the data. I understand that the response times vary by tissue, but is there a "foolproof" way to distinguish T1 from T2 by dicom data alone? In other words, is there a standard calculation that can be made on: (0018,0080) RepetitionTime (0018,0081) EchoTime In order to flag a series correctly, or do I need to rely on the series description .. (and if so, is the series description a re

Localizer Lines

Hi, i m Cantek =C3=87ET=C4=B0N. I want to ask specially to David CLUNIE. i m trying to add referance lines on my project but i couldnt get it. i read your writings but there are some points that i dont understand. I want to draw the lines on referance image so i need at least 2 points, but after your calculations i m getting only one and i think i m getting it wrong ! here is your calculations; dst_nrm_dircos_x =3D dst_row_dircos_y * dst_col_dircos_z - dst_row_dircos_z * dst_col_dircos_y; dst_nrm_dircos_y =3D dst_row_dircos_z * dst_col_dircos_x - dst_row_dircos_x * dst_col_d

Cmove Dicom code

Hi Everyone, I am newbie to dicom. I am trying to create a SCU that will establish an association with SCP and send a cmove request. I am pasting the method code for doing the cmove request in java. Can someone who is familiar or has done this please verify if my cmove code is correct? I would like to get some feedback on if I am missing any steps. Any feedback is appreciated. protected MCdimseMessage performCMove(final MCassociation a, final MCdataSet data) { MCdimseMessage rq = null; if (data != null) { rq = doSendServiceRequest(a, data); rq.setServiceCommand(MODALI

Worklist from RIS question

I have a question regarding scheduled procedures (we use RIS generated worklists) Patient "A" and "B" are scheduled for an exam and patient "A" does not show up for the exam, and reschedules for 5 days in the future, RIS uses the same Study UID (does not cancel the order, meerly reschedules the order). In the meantime the technologist has used Patient "A" scheduled exam by mistake on patient "B", tech realizes mistake and moves images to patient "B", technologist merges the correct order in PACS that it has for patient "B" (Study UID is one that was scheduled for patient "A") Patie

Call for Papers - DICOM 2008 International Conference & Seminar

DICOM 2010 INTERNATIONAL CONFERENCE & SEMINAR Call for Papers 9 - 11 October 2010 Rio de Janeiro, Brazil The DICOM 2010 International Conference and Seminar -- held in conjunction with the 2010 Brazilian Congress of Radiology -- will discuss the current, anticipated, and future features of the DICOM Standard and its role in health care. - Invited lectures will discuss current features of the DICOM standard and how they interplay with each other and the rest of the medical world. - The conference will include oral and poster presentations on new and upcoming DICOM features a

DICOM image transformations

Goodmorning, I need to ask if the transformations (change grayscale, rotation, contrast,..., or adding ROI or any other indication in DICOM file) is normally performed by viewers or any other program that manages DICOM, are normally write as tags or if image data are changed? Are there tags that can describe all the normal transformations on the image? If yes, viewers are able to display the correct transformations reading these tags? Thanks for any answer..

Transfering DICOM Images over network

Hi I am now in the initial phase of a teleradiology project. The radiologist have to review the case from any location in the world. The DICOM image may be in the PACS server and the radiologist have to view it through the DICOM Viewer. As the DICOM Images may be having larger file size , it will be little difficult to transfer over the network. We are thinking about the secured VPN for this . Is it a better one fro the transfer of the DICOM Images? Can any one suggest a better option for the DICOM Image transfer over network. Thanks in advance R.Joseph

How to start with gdcm

Hi, I am a complete newbie to gdcm. I want to use gdcm to extract some header data and a single image from dicom files. I thought it would be a good start to try out the HelloWorld example from the gdcm documentation. My problem is that I don't know what libraries or Dlls to include from the gdcm installation. I installed the precompiled version for windows. Can anyone point me to the right direction, please? thanks in advance Stefan

Looking for exhaustive list of "standard" hanging protocols, by modality

Hello, I work for a developer of radiological PACS and diagnostic viewing products. Our viewing software already has a decent method for customizing hanging protocols, but we're looking to make it a lot better. As part of our research, I'm hoping to find a list that details common (or popular) layout, or reading styles based on modality, body part, etc. i know this will be a massive list, and will vary greatly from radiologist to radiologist, and this is precisely part of our challenge. we have begun to interview our clients, but every scenario is so unique, we want to be su

Unique identification of patient

Hi, I am a newbie in RIS. When developing a RIS, we need to give an unique identification for each patient. I would like to know how this can be achieved. Can anybody help me. I have read in few articles that a MRN number is generated for each patient during the registration time. But what will happen if patients on different hospital have the same identification number? Hope you understood what i am asking. Thanks and regards, Sreelash

C-FIND RQ without matching keys

What's the right behavior of a DICOM C-FIND SCP (of the Composite Object Query/Retrieve Service Class, the Modality or the General- Purpose Worklist Management SOP Class , the Substance Administration Query Service Class, or the Hanging Protocol Query/Retrieve Service Class), receiving a C-FIND RQ with an identifier with only contains Key attributes without values? The question is raised by the sentence "If there are no matching keys, then there are no matches, return a response with a status equal to Success and with no Identifier." in the specification for such C-FIND SCPs. So

DICOM Printer

I am looking for a DICOM printer with support for almost all SOP classes. Can anyone please suggest Thanks and Regards, Sanjay

Interpreting DICOM Pixel Data Type.

How do you go about interpreting the VR of Pixel Data. Particularly in an implicit environment. There seem to be a number of ways to interpret these keys Bits Per Pixel Samples Per Pixel High Bit Bits Allocated Bits Stored Pixel Representation Photometric Interpretation. Where in the standard can I find a discussion of this. I looked through part 3 and found the elements.. but... Should I have been looking at the SOP Class definition? Would it differ between SOPs? Is there some way to tell if a 32 bit value is a float or a long integer.... Stuff like that....

DCM4CHEE suppports Presentation State SOP ?

I would like to send the image and the softcopy Presentation state to the PACS I am planning to use the dcm4chee as the image archieve. Before that i would like to confirm that : "Does dcm4chee support the Presentations state SOP classes ?" Thanks in advance Regards Robin

Significant change in pixelmed and dicom3tools: keywords updated per CP 850/2009

CP 850: "ftp://medical.nema.org/medical/dicom/final/cp850_ft.pdf" finally added rules for generating keywords for all data elements, and this was instantiated with defined keywords in the 2009 edition of PS 3.6. Accordingly, I have updated both dicom3tools and the pixelmed toolkit to use the new keywords in place of those used previously. This will affect both scripts that use the command line tools with keywords, as well as code that uses the keywords to look up tags in the dictionary. In the pixelmed toolkit, the database column names will also change in some cases.

Best Image Processing Open Source API in JAVA

I would like to create DICOM Image viewer and also want to implement DICOM services like C-FIND,C-MOVE etc to communicate with the PACS I would like to use the API for both the viewer and DICOM services. Could any one please tell me which is the best open source API in JAVA Thanks in advance Robin Joseph..

rtstruct series UID

Hi, I have a question regarding the series UID of dicom rtstructs. It seems to me that the series UID of an rtstruct should match the series UID from which the segmentation was made. For instance, if I segment off of a CT image, then the rtstruct that's produced should have the same series UID as that CT Image, but a different instance UID (to indicate that it's a different bit of information in the same series). I mention this because I'm trying to create rtstructs using this code here: http://www.insight-journal.org/browse/publication/316 However, they are creating their