Difference between revisions of "VolCT - Group 1A"

From QIBA Wiki
Jump to navigation Jump to search
Line 1: Line 1:
 +
==Current Goals & Status:==
 +
 +
* Goal: Benchmark intra/inter-reader variability for lung nodule volume measurements
 +
** Status: Most of Dataset acquired; finalizing reader software
 +
** Related Profile:
 +
 +
 +
== Work Documents ==
 
*[[Vol-CT - 1A Group Call Summaries]]
 
*[[Vol-CT - 1A Group Call Summaries]]
 
*[[Media:Qiba-nxp-ProposedPhase1A-2008-11-13-v05.ppt|Proposed 1A Protocol Outline (2008-11-13-v05)- For Comment from QIBA Members]]
 
*[[Media:Qiba-nxp-ProposedPhase1A-2008-11-13-v05.ppt|Proposed 1A Protocol Outline (2008-11-13-v05)- For Comment from QIBA Members]]
 +
 +
== Projects ==
 +
''May want to split these out to separate pages later''
 +
 +
===VolCT Lung Anthropomorphic Phantom Study===
 +
 +
====Objective: ====
 +
Measure intra- & inter-reader bias and variability phantom lesions for:
 +
:* Uni-dimensional size measurement
 +
:* Semi-automatic 3D volumetric measure
 +
 +
May compare with fully automated algorithm(s)
 +
 +
====Dataset:====
 +
Ground truth has been established by physical measurement "ex vivo“ on FDA phantom inserts.
 +
 +
Nodules (10 attached nodules)
 +
:* -10 & +100HU
 +
:* 10, 20 mm spheres
 +
:* 10 mm ovoid, lobulated, spiculated
 +
 +
Image Dataset
 +
:* 100 mAs exposures
 +
:* 0.75 & 5.0 mm slices
 +
:* 1 recon kernel
 +
:* Status:
 +
:** mostly acquired by FDA/CDRH/OSEL.
 +
:** Missing: 10 mm ovoid, spic, lob (Est: 12/01/08)
 +
 +
Acquisition Protocol:
 +
:* Scanner: Philips 16 slice
 +
:* Exposure (120 kVp): 100 mAs
 +
:* Slice thickness (50% overlap): 0.75 & 5.0 mm slices
 +
:* Recon kernel: Standard/medium (Still on table: Detail/Lung kernel)
 +
:* Pitch: 1.2
 +
:* 2 repeat scans
 +
:* 40 segmentations in set
 +
 +
The study is being conducted as a pilot.  The size (data, readers) has not been selected for any specific level of significance.
 +
 +
====Study Protocol:====
 +
Expert readers will measure/estimate nodule size from CT images.
 +
 +
Readers: 6 RadPharm radiologists
 +
 +
Software:
 +
:* ''Wendy will visit RadPharm and provide more info next week''
 +
:* In-house review software (Siemens?)
 +
:** Semi-automated 3D volume software
 +
:** Uni-dimensional measure (RESIST)
 +
:* ''Which fully automated software?''
 +
 +
Reading Session:
 +
:* Readers read all cases in 2 different reading sessions
 +
:** Random ordering
 +
:*** One-dimensional measure
 +
:*** Semi-automated segmentation
 +
:** Sessions separated by 3 week(?)
 +
:** Include duplicate cases within each read session (1/3-1/2 of cases for intra-reader estimates)
 +
:* Time restrictions:  Probably not (?)
 +
:* Specific instructions:  Probably not (?)
 +
 +
====Analysis:====
 +
Estimate intra- and inter-reader variability in the different volume estimate
 +
:* Estimate bias from known truth
 +
:* Estimate variability
 +
 +
Compare the bias and variability with the different methods
 +
 +
====Outstanding Issues:====

Revision as of 22:38, 11 December 2008

Current Goals & Status:

  • Goal: Benchmark intra/inter-reader variability for lung nodule volume measurements
    • Status: Most of Dataset acquired; finalizing reader software
    • Related Profile:


Work Documents

Projects

May want to split these out to separate pages later

VolCT Lung Anthropomorphic Phantom Study

Objective:

Measure intra- & inter-reader bias and variability phantom lesions for:

  • Uni-dimensional size measurement
  • Semi-automatic 3D volumetric measure

May compare with fully automated algorithm(s)

Dataset:

Ground truth has been established by physical measurement "ex vivo“ on FDA phantom inserts.

Nodules (10 attached nodules)

  • -10 & +100HU
  • 10, 20 mm spheres
  • 10 mm ovoid, lobulated, spiculated

Image Dataset

  • 100 mAs exposures
  • 0.75 & 5.0 mm slices
  • 1 recon kernel
  • Status:
    • mostly acquired by FDA/CDRH/OSEL.
    • Missing: 10 mm ovoid, spic, lob (Est: 12/01/08)

Acquisition Protocol:

  • Scanner: Philips 16 slice
  • Exposure (120 kVp): 100 mAs
  • Slice thickness (50% overlap): 0.75 & 5.0 mm slices
  • Recon kernel: Standard/medium (Still on table: Detail/Lung kernel)
  • Pitch: 1.2
  • 2 repeat scans
  • 40 segmentations in set

The study is being conducted as a pilot. The size (data, readers) has not been selected for any specific level of significance.

Study Protocol:

Expert readers will measure/estimate nodule size from CT images.

Readers: 6 RadPharm radiologists

Software:

  • Wendy will visit RadPharm and provide more info next week
  • In-house review software (Siemens?)
    • Semi-automated 3D volume software
    • Uni-dimensional measure (RESIST)
  • Which fully automated software?

Reading Session:

  • Readers read all cases in 2 different reading sessions
    • Random ordering
      • One-dimensional measure
      • Semi-automated segmentation
    • Sessions separated by 3 week(?)
    • Include duplicate cases within each read session (1/3-1/2 of cases for intra-reader estimates)
  • Time restrictions: Probably not (?)
  • Specific instructions: Probably not (?)

Analysis:

Estimate intra- and inter-reader variability in the different volume estimate

  • Estimate bias from known truth
  • Estimate variability

Compare the bias and variability with the different methods

Outstanding Issues:

Retrieved from "https://qibawiki.rsna.org/index.php?title=VolCT_-_Group_1A&oldid=342"