spie medical imaging conference lung imaging database consortium (lidc) lidc data elements and data...

46
SPIE Medical Imaging Conference SPIE Medical Imaging Conference Lung Imaging Database Consortium Lung Imaging Database Consortium (LIDC) (LIDC) LIDC Data Elements and Data LIDC Data Elements and Data Collection Process Collection Process February 13, 2005 February 13, 2005

Upload: amber-sherman

Post on 22-Dec-2015

227 views

Category:

Documents


0 download

TRANSCRIPT

SPIE Medical Imaging ConferenceSPIE Medical Imaging ConferenceLung Imaging Database Consortium (LIDC) Lung Imaging Database Consortium (LIDC)

LIDC Data Elements and Data Collection LIDC Data Elements and Data Collection ProcessProcess

February 13, 2005February 13, 2005

The The DatabaseDatabase•The Database will contain:The Database will contain:

1)1) A collection of CT scan imagesA collection of CT scan images

1

Low Dose Lung Cancer Screen CTsLow Dose Lung Cancer Screen CTs

2

3

4

5

The The DatabaseDatabase•CT scan image dataCT scan image data

a)a) Low Dose, Full Chest Lung Cancer Low Dose, Full Chest Lung Cancer Screening CTsScreening CTs

b)b)Conventional Dose, Full Chest CTs Conventional Dose, Full Chest CTs (Retrospective Cases, e.g. Lung (Retrospective Cases, e.g. Lung Ca Patients)Ca Patients)

c)c) Conventional Dose, Limited Chest Conventional Dose, Limited Chest CTs (e.g. Scan of a single nodule; CTs (e.g. Scan of a single nodule; biopsy)biopsy)

•All image data in DICOM formatAll image data in DICOM format•Indexing to identify cases of each Indexing to identify cases of each

typetype

The The DatabaseDatabase•The database will contain:The database will contain:

1)1) A collection of CT scan imagesA collection of CT scan images2)2) Technical factors about the CT Technical factors about the CT

scanscan

The The DatabaseDatabase•Technical factors about the CT scanTechnical factors about the CT scan

• All Image data will be anonymizedAll Image data will be anonymized• Much of DICOM header info will be Much of DICOM header info will be

preserved and stored along with image preserved and stored along with image datadata

• Database fields (can be queried) will Database fields (can be queried) will have Non-patient information from have Non-patient information from DICOM headerDICOM header

The The DatabaseDatabase•Technical factors about the CT scanTechnical factors about the CT scan

• Technical factors such as:Technical factors such as:• kVpkVp• Tube currentTube current• Rotation timeRotation time• ExposureExposure• Reconstructed slice thickness and Reconstructed slice thickness and

slice location (will still have to calc. slice location (will still have to calc. spacing)spacing)

• Reconstruction algorithmReconstruction algorithm• Reconstructed field of view, pixel sizeReconstructed field of view, pixel size

• Content in private fields WILL BE Content in private fields WILL BE removedremoved

The The DatabaseDatabase•The database will contain:The database will contain:

1)1) A collection of CT scan imagesA collection of CT scan images2)2) Technical factors about the CT Technical factors about the CT

scanscan3)3) Nodule Markings and Nodule Markings and

DescriptionsDescriptions

The The DatabaseDatabase•Nodule Markings and DescriptionsNodule Markings and Descriptions

1)1) For Nodules > 3 mm diameterFor Nodules > 3 mm diameter• Radiologist drawn boundariesRadiologist drawn boundaries• Description of characteristics (from defined Description of characteristics (from defined

list)list)

2)2) For Nodules < 3 mmFor Nodules < 3 mm• Radiologist marks only centroidRadiologist marks only centroid• No description characteristicsNo description characteristics

Nodule MarkingNodule MarkingInitial ApproachInitial Approach

Multiple Reads with Multiple ReadersMultiple Reads with Multiple Readers First Read – 4 readers, each reads First Read – 4 readers, each reads

independently (Blinded)independently (Blinded) Compile 4 blinded reads and distribute to Compile 4 blinded reads and distribute to

readersreaders Second Read – Same 4 readers, this time Second Read – Same 4 readers, this time

unblinded to the results of the other unblinded to the results of the other readers from the first reading.readers from the first reading.

No forced consensus on either location of No forced consensus on either location of nodules nor on their boundaries.nodules nor on their boundaries.

Blinded Reads – Each Reader Reads Independently (Blinded to Results of Other Readers)

Reader 1

Blinded Read for Reader 1 – Marks Only One Nodule

Reader 2

Blinded Read for Reader 2 – Marks Two Nodules(Note: One nodule is same as Reader 1)

Reader 3

Blinded Read for Reader 3 – Marks Two Nodules(Note: Again, One nodule is same as for Reader 1)

Reader 4

Blinded Read for Reader 4 – Did Not Mark Any Nodules

2nd Round - UnBlinded Reads Readings in Which Readers Are Shown Results of Other Readers

Each Reader Marks Nodules After Being Shown Results From Their Own and Other Readers’ Blinded Reads (Each Reader Decides to Include or Ignore).

Reader 1

Unblinded Read for Reader 1 – Now Marks Two Nodules(Originally only marked one)

Reader 2

Unblinded Read for Reader 2 – Still Marks Two Nodules(No Change)

Reader 3

Unblinded Read for Reader 3 – Now Marks Three Nodules(Originally only marked two)

Reader 4

Unblinded Read for Reader 4 – Now Marks Three Nodules(Originally did not mark any)

4/4 Markings

2/4 Markings 2/4 Markings

Results of Unblinded Reads from All Four Readers

We will capture one aspect of reader variability in this way

Case 5, Slice 19Case 5, Slice 19

Radiologist 1 - Method Radiologist 1 - Method 11

Radiologist 1 - Method Radiologist 1 - Method 22

Radiologist 1 - Method Radiologist 1 - Method 33

Radiologist 2 - Method Radiologist 2 - Method 11

Radiologist 2 - Method Radiologist 2 - Method 33

Radiologist 3 - Method Radiologist 3 - Method 11

Radiologist 3 - Method Radiologist 3 - Method 22

Radiologist 3 - Method Radiologist 3 - Method 33

Radiologist 4 - Method Radiologist 4 - Method 11

Radiologist 4 - Method Radiologist 4 - Method 22

Radiologist 4 - Method Radiologist 4 - Method 33

Radiologist 5 - Method Radiologist 5 - Method 11

Radiologist 5 - Method Radiologist 5 - Method 33

Probabilistic Description of Probabilistic Description of BoundaryBoundary

Apply Threshold if Apply Threshold if DesiredDesired

The The DatabaseDatabase•The database will contain:The database will contain:

1)1) A collection of CT scan imagesA collection of CT scan images2)2) Technical factors about the CT Technical factors about the CT

scanscan3)3) Nodule Markings and Nodule Markings and

DescriptionsDescriptions4)4) Pathology results or diagnosis Pathology results or diagnosis

information whenever availableinformation whenever available

Pathology InformationPathology Information

In those cases in which pathology is In those cases in which pathology is available, we will extract from available, we will extract from reports:reports:

Whether histology or cytology was Whether histology or cytology was performedperformed If histology, try to establish the cell type If histology, try to establish the cell type

according to WHO classificationsaccording to WHO classifications If cytology, establish whether it was If cytology, establish whether it was

benign or malignantbenign or malignant

Pathology InformationPathology Information

If no pathology, other diagnostic If no pathology, other diagnostic information may be substituted when information may be substituted when available (such as 2 years Dx F/U with available (such as 2 years Dx F/U with no change in radiographic no change in radiographic appearance).appearance).

If neither is available, then case will If neither is available, then case will be used for detection purposes only.be used for detection purposes only.

SummarySummary

LIDC Data ElementsLIDC Data Elements Image DataImage Data Technical FactorsTechnical Factors Radiologists’ Contours of Nodules Radiologists’ Contours of Nodules

Data Collection ProcessData Collection Process Diagnosis Information where availableDiagnosis Information where available