fidopt Interactive view
CT to PET, synthesis

NSCLC Radiogenomics, CT to FDG-PET

CT (source) and FDG-PET in body-weight SUV (target) of one patient's PET/CT session, the CT brought onto the PET grid through their shared scanner frame (NSCLC Radiogenomics collection).

10 test caseswhole bodysource CT, target PET
Look at the images
Read before comparing (6)
  • Case. A CT and an attenuation-corrected FDG-PET of one PET/CT session; prepare.py checks that both DICOM series share their FrameOfReferenceUID. Nothing is registered: the CT is resampled onto the PET grid with a cubic B-spline, air outside its field, and rounded to whole Hounsfield units.
  • Target. Body-weight SUV from the PET's rescale values and its DICOM dose, weight and decay information (AutoPET_CTPET_Syn.suv_factor). The mask is the body in the shared field of view (AutoPET_CTPET_Syn.body).
  • Motion. The PET and the CT have different acquisition durations, and the shared frame does not remove breathing motion between them. Tracer uptake is not set by CT anatomy alone.
  • Range. A dynamic range of 20 SUV puts PSNR and SSIM on the scale of the other FDG-PET synthesis datasets; it clips neither the target nor its errors.
  • Baseline. On the 10 test cases a constant at the target's median inside the mask scores, as the median over the cases and its range, an MAE of 0.47 (0.29 to 0.60) SUV, a PSNR of 24.0 (16.6 to 28.9) dB and an SSIM of 0.77 (0.73 to 0.85).
  • Overlap. MSD_Lung's 20 cases, other images (the diagnostic thin-section CTs), are of patients R01-004 to R01-081. MSD_Lung's 42 training cases not built and the decathlon's 32 unlabelled test scans were not matched to patients and may be of this pool. AMC is a separate cohort. SAROSBodyCT excludes the NSCLC Radiogenomics collection.

No method has a result on every case of this dataset yet.

Case by case

Click a score to look at that result: the target image, and over it the method's synthetic image. Underlined scores have their image; for the others you see the case without the result.

About the data

The first test case of NSCLC Radiogenomics, CT to FDG-PET
The first test caseClick to look at every case.
Test cases
10 test patients drawn with seed 2026 by cohort, R01 or AMC, then 10 development patients, one session each. The R01 pool starts at R01-082, after MSD_Lung's patients. prepare.py pins each selected patient's PET and CT series.
Truth
target image as released the attenuation-corrected FDG-PET of the same PET/CT session in body-weight SUV, aligned by the scanner frame
Licence
CC BY 3.0
10 cases for tuning, never tested

R01-128, AMC-040, R01-126, R01-100, AMC-004, R01-140, R01-117, AMC-032, R01-163, R01-087