I want to find the vector components of an MRI pixel -- How?

Click For Summary
SUMMARY

The discussion focuses on extracting vector components (x, y, z) of pixels from a DICOM image to analyze magnetic field vectors for H1 atoms, specifically with a frequency of 297 and a magnetic field strength (B0) of 3. Participants emphasize that DICOM images represent voxels that can be translated into T1 or T2 parameters, which are stored as integer values multiplied by a scaling factor found in the image header. Additional information regarding the spatial properties of the image is necessary to accurately derive a magnetic field map from pixel values.

PREREQUISITES
  • Understanding of DICOM image format and structure
  • Knowledge of MRI parameters, specifically T1 and T2 relaxation times
  • Familiarity with magnetic field concepts in MRI
  • Basic skills in programming for image processing (e.g., Python with NumPy)
NEXT STEPS
  • Research how to extract pixel values from DICOM images using libraries like pydicom
  • Learn about magnetic field mapping techniques in MRI
  • Explore the calculation of T1 and T2 parameters from MRI data
  • Investigate scaling factors in DICOM headers and their impact on pixel value interpretation
USEFUL FOR

Medical imaging professionals, MRI technologists, researchers in magnetic resonance imaging, and anyone involved in analyzing DICOM images for magnetic field studies.

mriphysics
Messages
1
Reaction score
0
I have a Dicom Image that is interpreted as an array. Each cell is a pixel. Each pixel has a value. I want to find the vector components (x,y,z) of a pixel. Specifically, I want to find magnetic field vectors for each pixel. The image is for H1 atoms. Freq = 297, B0 = 3.

How would you do this and can you show a sample calculation?
 
Biology news on Phys.org
I'm not really an MRI guy, but are you asking about a map of magnetic field inhomogeneities?

I would expect the voxels within a DICOM MRI image set to translate directly into a T1 or a T2 parameter or something like that. In DICOM format they're probably stored as some kind of integer value that gets multiplied by a scaling factor that's in the header. You would need more information to go from the pixel values to a magnetic field map, I would think. Is the image of a phantom with known, spatially uniform properties?
 

Similar threads

Replies
1
Views
3K
Replies
4
Views
3K
  • · Replies 3 ·
Replies
3
Views
397
  • · Replies 17 ·
Replies
17
Views
2K
Replies
13
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
Replies
19
Views
3K