Using DICOM and NIfTI in R
3 likes2,603 views
The document provides an overview of using DICOM and NIfTI data standards in R for manipulating, analyzing, and visualizing medical imaging data. It discusses the functionality of the oro.dicom and oro.nifti packages, covers DICOM file structure, and explains conversion processes from DICOM to NIfTI formats. Additionally, it addresses the advantages of using specific R packages for medical imaging data and includes references for further reading.
Using DICOM and NIfTI in R
- 1. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography DICOM and NIfTI Data Standards in R Brandon Whitcher PhD CStat Mango Solutions London, United Kingdom www.mango-solutions.com bwhitcher@mango-solutions.com @MangoImaging 12 June 2012 â useR!2012 Tutorial
- 2. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Outline 1 Summary 2 Digital Imaging and Communications in Medicine 3 Neuroimaging Informatics Technology Initiative 4 Data Visualization 5 DICOM-to-NIfTI Conversion 6 Conclusions
- 3. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Goal âą Content presented here provides users with the skills to manipulate DICOM / ANALYZE / NIfTI ïŹles in R. âą Real-world data sets are used to illustrate the basic functionality of oro.dicom and oro.nifti. âą S4 classes âniftiâ and âanlzâ enable further statistical analysis in R without losing contextual information from the original ANALYZE or NIfTI ïŹles. âą Images in the metadata-rich DICOM format may be converted to NIfTI semi-automatically using as much information from the DICOM ïŹles as possible. âą Basic visualization functions, are provided for âniftiâ and âanlzâ objects. âą The oro.nifti package allows one to track every operation on a nifti object in an XML-based audit trail.
- 4. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Alternatives to oro.dicom and oro.nifti âą There are several R packages that are able to access the DICOM / ANALYZE / NIfTI data formats: âą AnalyzeFMRI (Bordier et al. 2009) âą fmri (Polzehl and Tabelow 2007) âą Rniftilib (Granert 2010) âą tractor.base (Clayden 2010) Question #1 What are the (dis)advantages to having a single R package that performs input / output for medical imaging data? Question #2 Should R packages be discouraged from writing output in formats other than ANALYZE or NIfTI?
- 5. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography The DICOM âStandardâ âą The DICOM âstandardâ for data acquired using a clinical imaging device is very broad and complex. âą Each DICOM-compliant ïŹle is a collection of ïŹelds organized into two two-byte sequences (group,element) that are represented as hexadecimal numbers and form a tag. âą The (group,element) combination establishes what type of information is forthcoming in the ïŹle. âą There is no ïŹxed number of bytes for a DICOM header. âą The ïŹnal (group,element) tag should be the PixelData tag (7FE0,0010), such that all subsequent information is related to the image. DICOM Digital Imaging and Communications in Medicine http://medical.nema.org
- 6. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography The Structure of a DICOM File Data element with explicit VR of OB, OF, OW, SQ, UT or UN: +-----------------------------------------------------------+ | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | +----+----+----+----+----+----+----+----+----+----+----+----+ |<Group-->|<Element>|<VR----->|<0x0000->|<Length----------->|<Value-> Data element with explicit VR other than as shown above: +---------------------------------------+ | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | +----+----+----+----+----+----+----+----+ |<Group-->|<Element>|<VR----->|<Length->|<Value-> Data element with implicit VR: +---------------------------------------+ | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | +----+----+----+----+----+----+----+----+ |<Group-->|<Element>|<Length----------->|<Value-> âą Byte ordering for a single (group,element) tag in the DICOM standard. âą Explicit VRs store the VR as text characters in two bytes.
- 7. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Reading DICOM Files readDICOMFile() Accessing information stored in a single DICOM ïŹle âą The resulting object is a list with two elements: the DICOM header (hdr) and the DICOM image (img). âą Header information is organized in a data frame with six columns and an unknown number of rows. âą First ïŹve columns taken from DICOM header information (group, element, code, length and value) or inferred (name). âą (group,element) values are stored as character strings â not hexadecimal numbers. readDICOM() Accessing multiple DICOM ïŹles in a single directory or directory tree
- 8. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Stacking DICOM Images create3D() and create4D() Create arrays from DICOM headers / images âą Minimum input = âdcmâ structure âą PixelData may be read on-the-ïŹy âą Siemens MOSAIC format allowed âą 4D volumes may require additional information âą nslices = ? âą ntimes = ?
- 9. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography The ANALYZE Format âą The ANALYZE format was originally developed in conjunction with an image processing system (of the same name) at the Mayo Foundation. âą A common version of the format, although not the most recent, is called ANALYZE 7.5. âą An ANALYZE 7.5 format image is comprised of two ïŹles, the â.hdrâ and â.imgâ ïŹles, that contain information about the acquisition and the acquisition itself, respectively. The ANALYZE Format .hdr = 348-byte binary ïŹle of header information .img = binary ïŹat ïŹle of images (multi-dimensional array)
- 10. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography The NIfTI-1 Format âą The NIfTI-1 data format is almost identical to the ANALYZE format, but offers a few improvements: âą Merging of the header and image information into one ïŹle â.niiâ âą Re-organization of the 348-byte ïŹxed header into more relevant categories âą Possibility of extending the header information âą Discussions have begun on the NIfTI-2 data format. NIfTI Neuroimaging Informatics Technology Initiative http://nifti.nimh.nih.gov
- 11. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography S4 Classes for ANALYZE and NIfTI-1 âą âniftiâ âą Inherits from class âarrayâ âą Slots contain NIfTI header information âą Basic methods: show(), nifti(), is.nifti(), as(<obj>, âniftiâ) âą Input / output: readNIfTI(), writeNIfTI() âą Slot access: pixdim(), qform(), sform(), descrip(), aux.ïŹle(), audit.trail() âą Additional classes: âniftiExtensionâ and âniftiExtensionSectionâ âą âanlzâ âą Inherits from class âarrayâ âą Slots contain ANALYZE header information
- 12. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Visualization âą oro.nifti offers three functions for visualization âą image() = overloaded function for âanlzâ, âarrayâ and âniftiâ classes âą overlay() = extension of image() with x and y input parameters âą orthographic() = mid-axial, mid-sagittal, mid-coronal views Interactive Visualization FSLView http://www.fmrib.ox.ac.uk/fsl/fslview/ MRIcron http://www.cabiatl.com/mricro/mricron/ VolView http://www.kitware.com/products/volview.html
- 13. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography DICOM-to-NIfTI Conversion âą oro.dicom and oro.nifti were designed to use as much information as possible from the metadata-rich DICOM format and apply that information in the construction of the NIfTI data volume. âą Read in a single series using dicomSeperate() âą dicom2nifti() converts the list of DICOM images into a multidimensional âniftiâ object âą dicom2analyze() converts the list of DICOM images into a multidimensional âanlzâ object âą Additional scripting in R is required to deal with multiple series
- 14. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Goal âą Content presented here provides users with the skills to manipulate DICOM / ANALYZE / NIfTI ïŹles in R. âą Real-world data sets are used to illustrate the basic functionality of oro.dicom and oro.nifti. âą S4 classes âniftiâ and âanlzâ enable further statistical analysis in R without losing contextual information from the original ANALYZE or NIfTI ïŹles. âą Images in the metadata-rich DICOM format may be converted to NIfTI semi-automatically using as much information from the DICOM ïŹles as possible. âą Basic visualization functions, are provided for âniftiâ and âanlzâ objects. âą The oro.nifti package allows one to track every operation on a nifti object in an XML-based audit trail.
- 15. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Help âą http://rigorousanalytics.blogspot.com âą https://r-forge.r-project.org/projects/ rigorous âą rigorous-dicom@lists.r-forge.r-project.org âą rigorous-nifti@lists.r-forge.r-project.org âą rigorous-dcemris4@lists.r-forge.r-project.org Volume 44 in the Journal of Statistical Software Special volume on âMagnetic Resonance Imaging in Râ âą 13 articles on fMRI, DTI, DCE-MRI, etc. âą www.jstatsoft.org/v44
- 16. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Bibliography I C. Bordier, M. Dojat, and P. Lafaye de Micheaux. AnalyzeFMRI: an R package to perform statistical analysis on fmri datasets. 2009. URL http: //www.biostatisticien.eu/AnalyzeFMRI/. Software: R Package, AnalyzeFMRI, version 1.1-12. J. Clayden. tractor.base: A Package for Reading, Manipulating and Visualising Magnetic Resonance Images, 2010. URL http: //CRAN.R-project.org/package=tractor.base. R package version 1.5.0. O. Granert. Rniftilib: R Interface to NIFTICLIB (V1.1.0), 2010. URL http: //CRAN.R-project.org/package=Rniftilib. R package version 0.0-29.
- 17. DICOM and NIfTI Data Standards in R Summary DICOM NIfTI Visualization Conversion End Bibliography Bibliography II J. Polzehl and K. Tabelow. fmri: A package for analyzing fmri data. RNews, 7(2):13â17, 2007. URL http://www.r-project.org/doc/Rnews/Rnews_ 2007-2.pdf.
- 18. DICOM and NIfTI Data Standards in R Appendix R Package: oro.dicom âą Title: Rigorous - DICOM Input / Output âą Description: Data input/output functions for data that conform to the Digital Imaging and Communications in Medicine (DICOM) standard, part of the Rigorous Analytics bundle. âą Depends: R (>= 2.13.0), utils âą Suggests: hwriter, oro.nifti (>= 0.2.9) âą License: BSD âą URL: http://rigorousanalytics.blogspot.com
- 19. DICOM and NIfTI Data Standards in R Appendix R Package: oro.nifti âą Title: Rigorous - NIfTI+ANALYZE+AFNI Input / Output âą Description: Functions for the input/output and visualization of medical imaging data that follow either the ANALYZE, NIfTI or AFNI formats. This package is part of the Rigorous Analytics bundle. âą Depends: R (>= 2.13.0), bitops, graphics, grDevices, methods, utils âą Suggests: XML âą Imports: splines âą Enhances: fmri âą License: BSD âą URL: http://rigorousanalytics.blogspot.com