Skip to content

Quantitative Indices Extension

Andrey Fedorov edited this page May 28, 2014 · 23 revisions

Objective

Design a 3D Slicer extension for calculation of quantitative features within regions of interest.

Source code: https://github.com/QIICR/Slicer-QuantitativeIndicesExt

Specific Tasks

  • Develop tests and documentation.

Quantitative Indices

  • Mean
  • Variance
  • Minimum
  • Maximum
  • Volume
  • RMS - root-mean-square of grayscale values in region of interest
  • Peak - maximum average gray value that is calculated from a 1 cubic centimeter sphere placed within the region of interest
  • 1st Quartile - 25th percentile grayscale value in region of interest
  • Median - 50th percentile grayscale value in region of interest
  • 3rd Quartile - 75th percentile grayscale value in region of interest
  • Upper Adjacent - first gray value not greater than 1.5 times the interquartile range
  • Quarter 1 Distribution - percent of gray values that fall within the first quarter of the grayscale range within the region of interest
  • Quarter 2 Distribution - percent of gray values that fall within the second quarter of the grayscale range within the region of interest
  • Quarter 3 Distribution - percent of gray values that fall within the third quarter of the grayscale range within the region of interest
  • Quarter 4 Distribution - percent of gray values that fall within the last quarter of the grayscale range within the region of interest
  • Total Lesion Glycolysis (TLG) - tumor volume X mean grayscale value (SUV images)
  • Glycolysis Quarter 1 - lesion glycolysis calculated from the first quarter of the grayscale range within the region of interest (SUV images)
  • Glycolysis Quarter 2 - lesion glycolysis calculated from the second quarter of the grayscale range within the region of interest (SUV images)
  • Glycolysis Quarter 3 - lesion glycolysis calculated from the third quarter of the grayscale range within the region of interest (SUV images)
  • Glycolysis Quarter 4 - lesion glycolysis calculated from the last quarter of the grayscale range within the region of interest (SUV images)
  • SAM - standardized added metabolic activity. A background-corrected, partial volume independent version of TLG (SUV Images)
  • SAM Background - intermediate value for calculation of SAM. Estimates local background around region of interest (SUV Images)
  • TODO: add i-th percentile?

Codes that are used to encode the SUVbw ROI quantitative measurements

Code Coding scheme designator Code meaning Flag used by CLI Encoded in parameter file Units
R-00317 SRT Mean mean Mean_s g/ml
C1711260 UMLS Variance variance Variance_s (g/ml)^2
R-404FB SRT Minimum min Min_s g/ml
G-A437 SRT Maximum max Max_s g/ml
G-D705 SRT Volume volume Volume_s ml
C2347976 UMLS RMS rms RMS_s g/ml
250136 99PMP Peak Value Within ROI peak Peak_s g/ml
R-00319 SRT Median median Median_s g/ml
250137 99PMP 25th Percentile Value quart1 First_Quartile_s g/ml
250138 99PMP 75th Percentile Value quart3 Third_Quartile_s g/ml
250139 99PMP Upper Adjacent Value adj Upper_Adjacent_s g/ml
250140 99PMP Percent Within First Quarter of Intensity Range q1 Q1_Distribution_s %
250141 99PMP Percent Within Second Quarter of Intensity Range q2 Q2_Distribution_s %
250142 99PMP Percent Within Third Quarter of Intensity Range q3 Q3_Distribution_s %
250143 99PMP Percent Within Fourth Quarter of Intensity Range q4 Q4_Distribution_s %
250144 99PMP Total Lesion Glycolysis (TLG) tlg TLG_s g
250145 99PMP Glycolysis Within First Quarter of Intensity Range gly1 Glycolysis_Q1_s g
250146 99PMP Glycolysis Within Second Quarter of Intensity Range gly2 Glycolysis_Q2_s g
250147 99PMP Glycolysis Within Third Quarter of Intensity Range gly3 Glycolysis_Q3_s g
250148 99PMP Glycolysis Within Fourth Quarter of Intensity Range gly4 Glycolysis_Q4_s g
250149 99PMP Standardized Added Metabolic Activity (SAM) sam SAM_s g
250150 99PMP Standardized Added Metabolic Activity (SAM) Background sambg SAM_Background_s g/ml

Also consider encoding measures computed by the following ITK filters:

Uptake Regions

  • Primary tumor
  • Lymph nodes
  • Reference regions - used for normalization
  • liver: spherical region
  • cerebellum - region or the whole thing
  • aortic arch (part of it, something where reader has confidence in, independently on the motion); may need to look at CT data to perform the measurement, but segmentation is done on PET
  • in clinical practice opposite side can be used as reference (not used in Iowa project)

Codes for uptake regions:

Code Coding scheme designator Code meaning
T-62000 SRT Liver
T-C4000 SRT Lymph node
T-A6000 SRT Cerebellum
T-42300 SRT Aortic arch
M-8FFFF SRT Neoplasm

Information we might want / need to encode:

  • Reader information
  • Site of the lesion/lymph node: primary cancer location is encoded in the database, but not the locations of lymph nodes
  • Tool/method used for contouring
  • Time needed for segmentation
  • Iteration segmentation was done (each structure is segmented multiple times)

Contributors

  • Ethan Ulrich
  • Markus Van Tol
  • Reinhard R. Beichel
  • Andrey Fedorov

References

  1. Mertens J, Dobbeleir A, Ham H, D’Asseler Y, Goethals I, Van de Wiele C. Standardized added metabolic activity (SAM): a partial volume independent marker of total lesion glycolysis in liver metastases. Eur J Nucl Med Mol Imaging. 2012;39(9):1441–8 PDF (private folder of the QIICR Midas community - login required)
  2. Orlhac, F., Soussan, M., Maisonobe, J.-A., Garcia, C. a, Vanderlinden, B., & Buvat, I. (2014). Tumor Texture Analysis in 18F-FDG PET: Relationships Between Texture Parameters, Histogram Indices, Standardized Uptake Values, Metabolic Volumes, and Total Lesion Glycolysis. Journal of Nuclear Medicine : Official Publication, Society of Nuclear Medicine. doi:10.2967/jnumed.113.129858 (http://www.ncbi.nlm.nih.gov/pubmed/24549286)
  3. Abd El-Hafez, Y. G., Moustafa, H. M., Khalil, H. F., Liao, C.-T., & Yen, T.-C. (2013). Total lesion glycolysis: a possible new prognostic parameter in oral cavity squamous cell carcinoma. Oral Oncology, 49(3), 261–8. doi:10.1016/j.oraloncology.2012.09.005
You can’t perform that action at this time.