PET/CT Segmentation Report

Case: ModelingLymphoma • Generated from Ingensia PET/CT sample pipeline run

Run date: 2025-10-03 10:25:38 GMT

1. Patient & Scan

Patient MRN	PETCT_fe705ea1cc
Patient name	PETCT_fe705ea1cc
Sex	F
Age	066Y
DOB	—

Exam date	2002-12-29
Institution	—
Accession	—
Study ID	—

2. Data Completeness

identifiers.json
Present

reconstruction_method.json
Present

dicom_scoring.json
Present

qc_report.txt
Present

geometry_report_pet.txt
Present

geometry_report_seg.txt
Present

dicom_detection.summary
Present

Final segmentation (BASENAME_SEG.nii.gz)
Present

Modeling features CSV
Present

3. DICOM Series Discovery & Selection

Series folder	Modality	Series description	Kernel	Recon method	CT score	PET score	SOP class	Radionuclide tags?
1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548.dicom_header Selected CT	CT	GK p.v.1 WF	B30f	—	9	0	CT Image Storage	No
1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058.dicom_header Selected PET	PT	PET corr.	XYZ Gauss2.00	—	2	15	Positron Emission Tomography Image Storage	Yes

Show all discovered series

Series folder	Modality	Series description	Kernel	Recon method	CT score	PET score	SOP class	Radionuclide tags?
1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898.dicom_header	SEG	Segmentation	—	—	0	0	Segmentation Storage	No
1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548.dicom_header Selected CT	CT	GK p.v.1 WF	B30f	—	9	0	CT Image Storage	No
1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058.dicom_header Selected PET	PT	PET corr.	XYZ Gauss2.00	—	2	15	Positron Emission Tomography Image Storage	Yes

4. Overview MIPs

5. Live DICOM viewer

Live DICOM viewer (auto-loaded demo) Open full screen ⤴

6. Modeling Visuals

7. Segmentation Metrics

n_voxels	580540
MTV_mL	226.2166495789401
TLG	1008.0412046735794
SUV_mean	4.456087589263916
SUV_median	3.682021141052246
SUV_max	19.534645080566406
SUV_peak	14.80296802520752
SUV_std	2.369410514831543

SUV_CoV	0.5317244033847496
SUV_p10	2.3759382486343386
SUV_p90	7.734170770645143
shape_Sphericity	0.22
shape_MeshVolume	225499.97
shape_MajorAxisLength	863.49
shape_MinorAxisLength	207.14
shape_LeastAxisLength	101.32

Show all radiomics features

Identifier	ModelingLymphoma
Segmentation_Model	Ingensia_PETAI_v2_Lymphoma_Melanoma
Date_Modeled	2025-10-03
Modeling_Time_Seconds	472
Modeling_Time_Hours	0.13
CT_Image_Reconstruction_Method	UNAVAILABLE
PET_Image_Reconstruction_Method	PSF+TOF 2i21s
CT_Image_Corrected	UNAVAILABLE
PET_Image_Corrected	NORM\|DTIM\|ATTN\|SCAT\|DECY\|RAN
CT_Image_Attenuation_Correction_Method	UNAVAILABLE
PET_Image_Attenuation_Correction_Method	measured;GK p.v.3 eFoV
CT_Image_Convolution Kernel	B30f
PET_Image_Convolution Kernel	XYZ Gauss2.00
CT_Reconstruction_Algorithm	UNAVAILABLE
CT_Reconstruction_Diameter	5.0\|-211.0\|-331.0
CT_Slice_Thickness	True
CT_Pixel_Spacing	0.74609375\|0.74609375
CT_Series_Description	GK p.v.1 WF
CT_Protocol_Name	04_PETCT_GK_pv_TH_Insp
CT_Convolution_Kernel_Group	UNAVAILABLE
PET_Randoms_Correction_Method	START
PET_Detector_Normalization_Method	UNAVAILABLE
PET_Scatter_Correction_Method	Model-based; relative scatter scaling
PET_Decay_Correction	UNAVAILABLE
PET_Units	BQML
PET_Frame_Reference_Time	UNAVAILABLE
PET_Actual_Frame_Duration	120000.0
Patient_Weight	51
Radiopharmaceutical_Start_Time	115000
Radionuclide_Total_Dose	308000000
Radionuclide_Half_Life	6586.2
PET_Series_Description	PET corr.
PET_Protocol_Name	UNAVAILABLE
CT_Reconstruction_Group	Unknown
PET_Reconstruction_Group	OSEM+PSF+TOF
Empty_Segmentation	False
n_voxels	580540
MTV_mL	226.2166495789401
SUV_mean	4.456087589263916
SUV_max	19.534645080566406
SUV_peak	14.80296802520752
SUV_median	3.682021141052246
SUV_std	2.369410514831543
SUV_CoV	0.5317244033847496
SUV_p10	2.3759382486343386
SUV_p25	2.832846701145172
SUV_p50	3.682021021842957
SUV_p75	5.232248425483704
SUV_p90	7.734170770645143
SUV_skew	1.7498685121536257
SUV_kurtosis	3.1790571212768555
TLG	1008.0412046735794
shape_Elongation	0.24
shape_Flatness	0.12
shape_LeastAxisLength	101.32
shape_MajorAxisLength	863.49
shape_Maximum2DDiameterColumn	665.31
shape_Maximum2DDiameterRow	645.55
shape_Maximum2DDiameterSlice	247.59
shape_Maximum3DDiameter	693.27
shape_MeshVolume	225499.97
shape_MinorAxisLength	207.14
shape_Sphericity	0.22
shape_SurfaceArea	81646.15
shape_SurfaceVolumeRatio	0.36
shape_VoxelVolume	226216.65
firstorder_10Percentile	-44
firstorder_90Percentile	121
firstorder_Energy	8255889774
firstorder_Entropy	3.47
firstorder_InterquartileRange	73
firstorder_Kurtosis	30.73
firstorder_Maximum	2746
firstorder_MeanAbsoluteDeviation	62.54
firstorder_Mean	48.11
firstorder_Median	71
firstorder_Minimum	-1024
firstorder_Range	3770
firstorder_RobustMeanAbsoluteDeviation	32.55
firstorder_RootMeanSquared	119.25
firstorder_Skewness	-3.22
firstorder_TotalEnergy	3217038833
firstorder_Uniformity	0.13
firstorder_Variance	11906.4
glcm_Autocorrelation	1912.08
glcm_ClusterProminence	128610.73
glcm_ClusterShade	-1820.98
glcm_ClusterTendency	65.2
glcm_Contrast	3.63
glcm_Correlation	0.89
glcm_DifferenceAverage	1.08
glcm_DifferenceEntropy	1.94
glcm_DifferenceVariance	2.43
glcm_Id	0.63
glcm_Idm	0.6
glcm_Idmn	1.0
glcm_Idn	0.99
glcm_Imc1	-0.3
glcm_Imc2	0.93
glcm_InverseVariance	0.46
glcm_JointAverage	43.55
glcm_JointEnergy	0.04
glcm_JointEntropy	5.78
glcm_MCC	0.92
glcm_MaximumProbability	0.12
glcm_SumAverage	87.1
glcm_SumEntropy	4.34
glcm_SumSquares	17.21
gldm_DependenceEntropy	7.51
gldm_DependenceNonUniformity	32756.03
gldm_DependenceNonUniformityNormalized	0.06
gldm_DependenceVariance	25.79
gldm_GrayLevelNonUniformity	77160.06
gldm_GrayLevelVariance	19.14
gldm_HighGrayLevelEmphasis	1906.51
gldm_LargeDependenceEmphasis	117.75
gldm_LargeDependenceHighGrayLevelEmphasis	233687.31
gldm_LargeDependenceLowGrayLevelEmphasis	0.06
gldm_LowGrayLevelEmphasis	0.0
gldm_SmallDependenceEmphasis	0.05
gldm_SmallDependenceHighGrayLevelEmphasis	81.07
gldm_SmallDependenceLowGrayLevelEmphasis	0.0
glrlm_GrayLevelNonUniformity	44298.71
glrlm_GrayLevelNonUniformityNormalized	0.11
glrlm_GrayLevelVariance	24.33
glrlm_HighGrayLevelRunEmphasis	1877.82
glrlm_LongRunEmphasis	3.25
glrlm_LongRunHighGrayLevelEmphasis	6310.6
glrlm_LongRunLowGrayLevelEmphasis	0.0
glrlm_LowGrayLevelRunEmphasis	0.0
glrlm_RunEntropy	4.97
glrlm_RunLengthNonUniformity	207946.85
glrlm_RunLengthNonUniformityNormalized	0.53
glrlm_RunPercentage	0.67
glrlm_RunVariance	0.98
glrlm_ShortRunEmphasis	0.76
glrlm_ShortRunHighGrayLevelEmphasis	1400.99
glrlm_ShortRunLowGrayLevelEmphasis	0.0
glszm_GrayLevelNonUniformity	839.95
glszm_GrayLevelNonUniformityNormalized	0.04
glszm_GrayLevelVariance	119.24
glszm_HighGrayLevelZoneEmphasis	1552.09
glszm_LargeAreaEmphasis	197236.62
glszm_LargeAreaHighGrayLevelEmphasis	391576186.93
glszm_LargeAreaLowGrayLevelEmphasis	99.7
glszm_LowGrayLevelZoneEmphasis	0.0
glszm_SizeZoneNonUniformity	6503.04
glszm_SizeZoneNonUniformityNormalized	0.3
glszm_SmallAreaEmphasis	0.57
glszm_SmallAreaHighGrayLevelEmphasis	859.96
glszm_SmallAreaLowGrayLevelEmphasis	0.0
glszm_ZoneEntropy	7.92
glszm_ZonePercentage	0.04
glszm_ZoneVariance	196520.04
ngtdm_Busyness	4.01
ngtdm_Coarseness	0.0
ngtdm_Complexity	4659.37
ngtdm_Contrast	0.0
ngtdm_Strength	1.0

8. Modeling Process Checks

QC passed	Yes
PET detection	meta_data -- PET PASS
CT↔PET geometry pass	True
CT↔SEG geometry pass	True
SEG labels valid	True

View QC report

========================================================================
 QC PASSED?  ✅  YES
========================================================================

QC details:
   PET_PASS            : PASS
   MIPS_PNGs           : PASS
   Geometry_PET_OK     : PASS
   Geometry_SEG_OK     : PASS
   SEG_file_present    : PASS

Segmentation status:
   Tumour found?      YES

Metrics (zeros shown for missing files):
   TMTV (mL)        : 226.2
   MeshVolume (mm³) : 225500.0
   SurfaceArea (mm²): 81646.2

View dicom_scoring.json

{
  "series": [
    {
      "file": "1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898.dicom_header",
      "score_ct": 0,
      "score_pet": 0,
      "tags": {
        "sop_class": "Segmentation Storage",
        "modality": "SEG",
        "series_description": "Segmentation",
        "image_type": [
          "DERIVED",
          "PRIMARY"
        ],
        "convolution_kernel": null,
        "radionuclide_tags_present": false,
        "reconstruction_diameter": null,
        "reconstruction_method": null
      }
    },
    {
      "file": "1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548.dicom_header",
      "score_ct": 9,
      "score_pet": 0,
      "tags": {
        "sop_class": "CT Image Storage",
        "modality": "CT",
        "series_description": "GK p.v.1 WF",
        "image_type": [
          "ORIGINAL",
          "PRIMARY",
          "AXIAL",
          "CT_SOM5 SPI"
        ],
        "convolution_kernel": "B30f",
        "radionuclide_tags_present": false,
        "reconstruction_diameter": null,
        "reconstruction_method": null
      }
    },
    {
      "file": "1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058.dicom_header",
      "score_ct": 2,
      "score_pet": 15,
      "tags": {
        "sop_class": "Positron Emission Tomography Image Storage",
        "modality": "PT",
        "series_description": "PET corr.",
        "image_type": [
          "ORIGINAL",
          "PRIMARY"
        ],
        "convolution_kernel": "XYZ Gauss2.00",
        "radionuclide_tags_present": true,
        "reconstruction_diameter": null,
        "reconstruction_method": null
      }
    }
  ],
  "master_ct": "1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548.dicom_header",
  "master_pet": "1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058.dicom_header"
}

9. Geometry Summaries

CT ↔ PET

Geometry pass	True
Segmentation labels	n/a (no SEG provided)
origin_match	True
spacing_match	True
direction_match	True
size_match	True

CT ↔ SEG

Geometry pass	True
Segmentation labels	True
origin_match	True
spacing_match	True
direction_match	True
size_match	True

Raw geometry_report_pet.txt

[1mProvided modalities:[0m CT: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT.nii.gz, PET: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET_resampled_bspline.nii.gz

[1mReading CT[0m

[1mReading PET[0m

[1m--- Geometry and metadata ---[0m
CT: Size (512, 512, 1214)  Spacing (0.74609375, 0.74609375, 0.70001220703125)  Origin (-185.626953125, -20.373046875, -1180.11474609375)
PET: Size (512, 512, 1214)  Spacing (0.74609375, 0.74609375, 0.70001220703125)  Origin (-185.626953125, -20.373046875, -1180.11474609375)

[1mGeometry CT ↔ PET[0m
origin_match: True
spacing_match: True
direction_match: True
size_match: True

[32m============================ Summary ============================[0m
Geometry pass:          True
Segmentation labels:    n/a (no SEG provided)
[32m[1m
✔ The supplied data appear nnU‑Net‑compatible.[0m

Raw geometry_report_seg.txt

[1mProvided modalities:[0m CT: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/preprocessed/CT.nii.gz, SEG: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/PETAI_001.nii.gz

[1mReading CT[0m

[1mReading SEG[0m

[1m--- Geometry and metadata ---[0m
CT: Size (512, 512, 1214)  Spacing (0.74609375, 0.74609375, 0.70001220703125)  Origin (-185.626953125, -20.373046875, -1180.11474609375)
SEG: Size (512, 512, 1214)  Spacing (0.74609375, 0.74609375, 0.70001220703125)  Origin (-185.626953125, -20.373046875, -1180.11474609375)

[1mGeometry CT ↔ SEG[0m
origin_match: True
spacing_match: True
direction_match: True
size_match: True

[1m--- Segmentation label checks ---[0m
unique_labels: [0, 1]
background_is_zero: True
all_integers: True
consecutive_labels: True

[32m============================ Summary ============================[0m
Geometry pass:          True
Segmentation labels:    True
[32m[1m
✔ The supplied data appear nnU‑Net‑compatible.[0m

[INFO] Logging to /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/ModelingLymphoma_report/log.txt [INFO] Computation started at: 2025-10-03 10:25:38 GMT ============================================================ [SECTION] Environment setup ============================================================ [STEP] Load optional host env (if present) [SUCCESS] Load optional host env (if present) ============================================================ [SECTION] Sanitize series layout ============================================================ [STEP] Sanitize series layout in /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma [INFO] Scanning recursively: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma [INFO] Discovered 3 series; will move 1499 files into series directories under root. [SUCCESS] Wrote manifest: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/sanitizer_manifest.csv [SUCCESS] Removed 2 empty directories. [SUCCESS] Series directories created under root (named by SeriesInstanceUID). [SUCCESS] Sanitize series layout in /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma ============================================================ [SECTION] DICOM header capture & scoring ============================================================ [STEP] Capture DICOM headers Wrote header for '1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058' -> /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058.dicom_header Error reading DICOM file /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/ModelingLymphoma_report/log.txt: File is missing DICOM File Meta Information header or the 'DICM' prefix is missing from the header. Use force=True to force reading. Wrote header for '1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548' -> /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548.dicom_header Warning: no files found in /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/34234234, skipping. Wrote header for '1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898' -> /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898.dicom_header Error reading DICOM file /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898.dicom_header: File is missing DICOM File Meta Information header or the 'DICM' prefix is missing from the header. Use force=True to force reading. Warning: no files found in /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/300.000000-Segmentation-09898, skipping. Warning: no files found in /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/982343, skipping. [SUCCESS] Capture DICOM headers [STEP] Determine reconstruction method [SUCCESS] Determine reconstruction method [STEP] Full DICOM header scoring [SUCCESS] Full DICOM header scoring [STEP] Detection summary scoring [SUCCESS] Detection summary scoring [STEP] Rename CT/PET sub-directories ============================================================ [PROCESSING STUDY] /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma ============================================================ [STEP] Verifying PET PASS status … [SUCCESS] PET PASS confirmed. [STEP] Extracting master_ct and master_pet values … [SUCCESS] Found CT UID: 1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548 [SUCCESS] Found PET UID: 1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058 [STEP] Generating rename commands … mv "/home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/1.3.6.1.4.1.14519.5.2.1.4219.6651.236102552014587977554483752548" "/home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT" mv "/home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/1.3.6.1.4.1.14519.5.2.1.4219.6651.484574363818075625229557398058" "/home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET" [STEP] Executing rename commands … [SUCCESS] Rename complete. [SUCCESS] Rename CT/PET sub-directories [STEP] Create CT and PET specific dicom_header files (from INPUT → META_DIR) Wrote header for 'CT' -> /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/CT.dicom_header Wrote header for 'PET' -> /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/PET.dicom_header [SUCCESS] Create CT and PET specific dicom_header files (from INPUT → META_DIR) [STEP] Create PET CT reconstruction csv detailing the reconstruction method Wrote: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/reconstruction.csv [SUCCESS] Create PET CT reconstruction csv detailing the reconstruction method ============================================================ [SECTION] NIfTI conversion & resampling ============================================================ [STEP] Convert CT DICOM → NIfTI (staging) ============================================================ [PROCESSING SERIES] /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT ============================================================ [STEP] Checking for dcm2niix … [STEP] Ensuring output directory exists → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma [STEP] Running dcm2niix … Chris Rorden's dcm2niiX version v1.0.20211006 (JP2:OpenJPEG) GCC11.2.0 x86-64 (64-bit Linux) Found 1214 DICOM file(s) Convert 1214 DICOM as /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT (512x512x1214x1) Conversion required 28.177000 seconds (28.169077 for core code). [SUCCESS] dcm2niix completed without errors. [SUCCESS] Wrote /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT.nii.gz [SUCCESS] Convert CT DICOM → NIfTI (staging) [STEP] Convert PET DICOM → NIfTI (staging) ============================================================ [PROCESSING SERIES] /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET ============================================================ [STEP] Checking for dcm2niix … [STEP] Ensuring output directory exists → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma [STEP] Running dcm2niix … Chris Rorden's dcm2niiX version v1.0.20211006 (JP2:OpenJPEG) GCC11.2.0 x86-64 (64-bit Linux) Found 284 DICOM file(s) Convert 284 DICOM as /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET (400x400x284x1) Saving as 32-bit float (slope, intercept or bits allocated varies). Conversion required 4.153035 seconds (4.151574 for core code). [SUCCESS] dcm2niix completed without errors. [SUCCESS] Wrote /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET.nii.gz [SUCCESS] Convert PET DICOM → NIfTI (staging) [STEP] Resample PET to CT grid (staging) Reading CT from /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/CT.nii.gz ... Reading PET from /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET.nii.gz ... Resampling PET to match CT geometry (B‑spline) ... ✓ PET saved to /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET_resampled_bspline.nii.gz Writing resampling details to /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/resampling.txt Resampling workflow complete! [SUCCESS] Resample PET to CT grid (staging) ============================================================ [SECTION] Geometry checks & MIPs (pre-modeling) ============================================================ [STEP] Geometry report (PET) [SUCCESS] Geometry report (PET) [STEP] Create CT+PET MIPs [INFO] Saved CT_PET_MIPS-like_coronal.png → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/CT_PET_MIPS-like_coronal.png [INFO] Saved CT_PET_MIPS-like_sagittal.png → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/CT_PET_MIPS-like_sagittal.png [SUCCESS] Create CT+PET MIPs ============================================================ [SECTION] Identifier & header extraction ============================================================ [STEP] Extract DICOM identifiers 2025-10-03 10:26:55,282 | INFO | Using header file: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/1.3.6.1.4.1.14519.5.2.1.4219.6651.163390901989343459255915209898.dicom_header 2025-10-03 10:26:55,283 | INFO | Parsing header … 2025-10-03 10:26:55,284 | WARNING | Missing fields: dob, institution, accession, study_id 2025-10-03 10:26:55,290 | INFO | Identifiers written to /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/identifiers.json [SUCCESS] Extract DICOM identifiers [STEP] Extract PET header for modeling (staging) Header extracted from: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/PET/1-001.dcm Saved to: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/PET.dicom_header [SUCCESS] Extract PET header for modeling (staging) ============================================================ [SECTION] Modeling directory preparation ============================================================ [STEP] Create modeling/input_files [SUCCESS] Create modeling/input_files [STEP] Create modeling/preprocessed [SUCCESS] Create modeling/preprocessed [STEP] Move preprocessed files to modeling/preprocessed (from staging) [SUCCESS] Move preprocessed files to modeling/preprocessed (from staging) [STEP] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/input_files [SUCCESS] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/input_files [STEP] Symlink CT to nnU-Net slot 0 [SUCCESS] Symlink CT to nnU-Net slot 0 [STEP] Symlink PET to nnU-Net slot 1 [SUCCESS] Symlink PET to nnU-Net slot 1 [STEP] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma [SUCCESS] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma ============================================================ [SECTION] Segmentation prediction ============================================================ [STEP] Validate nnU-Net environment (paths) [INFO] nnU-Net paths OK: nnUNet_raw = /dev/shm/ingensia_models.8953/nnU-Net_dirs/nnUNet_raw nnUNet_preprocessed= /dev/shm/ingensia_models.8953/nnU-Net_dirs/nnUNet_preprocessed nnUNet_results = /dev/shm/ingensia_models.8953/nnU-Net_dirs/nnUNet_results [SUCCESS] Validate nnU-Net environment (paths) [INFO] [TIMER] MODEL_START: 1759487215 [STEP] Run nnU-Net-v2 prediction step 1 ####################################################################### Please cite the following paper when using nnU-Net: Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2), 203-211. ####################################################################### There are 1 cases in the source folder I am process 0 out of 1 (max process ID is 0, we start counting with 0!) There are 1 cases that I would like to predict Predicting PETAI_001: perform_everything_on_device: True 100%|██████████| 8/8 [00:02<00:00, 3.96it/s] 100%|██████████| 8/8 [00:01<00:00, 5.15it/s] 100%|██████████| 8/8 [00:01<00:00, 5.15it/s] 100%|██████████| 8/8 [00:01<00:00, 5.14it/s] 100%|██████████| 8/8 [00:01<00:00, 5.14it/s] sending off prediction to background worker for resampling and export done with PETAI_001 real 1m26.038s user 1m39.175s sys 0m13.967s step 2 ####################################################################### Please cite the following paper when using nnU-Net: Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2), 203-211. ####################################################################### There are 1 cases in the source folder I am process 0 out of 1 (max process ID is 0, we start counting with 0!) There are 1 cases that I would like to predict Predicting PETAI_001: perform_everything_on_device: True 100%|██████████| 216/216 [00:53<00:00, 4.04it/s] 100%|██████████| 216/216 [00:53<00:00, 4.06it/s] 100%|██████████| 216/216 [00:53<00:00, 4.06it/s] 100%|██████████| 216/216 [00:53<00:00, 4.06it/s] 100%|██████████| 216/216 [00:53<00:00, 4.06it/s] sending off prediction to background worker for resampling and export done with PETAI_001 real 6m9.239s user 4m27.933s sys 2m9.795s step 3 step 4 real 0m3.012s user 0m26.617s sys 0m1.375s completed [SUCCESS] Run nnU-Net-v2 prediction [INFO] [TIMER] MODEL_END: 1759487687 [INFO] Modeling elapsed: 0h 7m 52s (472 s) ============================================================ [SECTION] Segmentation mask sanity check ============================================================ ============================================================ [SECTION] Geometry & MIPs (post-modeling) ============================================================ [STEP] Geometry report (SEG) [SUCCESS] Geometry report (SEG) [STEP] Create CT+SEG MIPs [INFO] Saved CT_SEG_MIPS-like_coronal.png → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/CT_SEG_MIPS-like_coronal.png [INFO] Saved CT_SEG_MIPS-like_sagittal.png → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/meta_data/CT_SEG_MIPS-like_sagittal.png [SUCCESS] Create CT+SEG MIPs ============================================================ [SECTION] Radiomics & SUV conversion ============================================================ [STEP] PyRadiomics feature extraction [2025-10-03 10:35:39] W: radiomics.glcm: GLCM is symmetrical, therefore Sum Average = 2 * Joint Average, only 1 needs to be calculated [SUCCESS] PyRadiomics feature extraction [STEP] PyRadiomics file conversion ✓ Wrote /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/radiomics_features.csv [SUCCESS] PyRadiomics file conversion [STEP] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/preprocessed [SUCCESS] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/preprocessed [STEP] Convert PET to SUV Header resolution report Field | Value | Units | Source | Required | Status | Note -------------- | ---------------------- | -------- | ---------- | -------- | -------- | -------------------------- Weight | 51 | kg | header | yes | OK | Dose | 308 | MBq | header | yes | OK | Half-life | 6586.2 | s | header | no | OK | Inj DateTime | 2002-12-29 11:50:00 | | header | no | OK | Scan DateTime | 2002-12-29 12:50:02 | | header | no | OK | Units | BQML | | header | no | OK | scaled to kBq/mL Rescale Slope | 1.0542 | | header | no | OK | Rescale Inter | 0 | | header | no | OK | Decay correction will be applied (Δt = 60.0 min). All required fields resolved. Proceeding with SUV conversion… SUV map saved: SUV.nii.gz min / median / max SUV = -0.482 / 0.002 / 50.326 NIfTI descrip: SUVbw wt=51.0kg dose0=308.0MBq doseCorr=210.8MBq dT=60.0min HL=6586s [SUCCESS] Convert PET to SUV [STEP] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling [SUCCESS] cd /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling [STEP] Compute SUV metrics (v2) ----- Patient‑level summary ----- TMTV_mL : 226.217 TLG_total : 1008.041 global_SUVmax : 19.535 Dmaxpatient_mm : 0.000 n_lesions : 1 [SUCCESS] Compute SUV metrics (v2) ============================================================ [SECTION] Write modeling_details.csv ============================================================ [STEP] Emit modeling_details.csv Wrote /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/modeling_details.csv [SUCCESS] Emit modeling_details.csv ============================================================ [SECTION] Schema-driven feature finalization ============================================================ [STEP] Build canonical modeling features (schema-aligned) Wrote canonical modeling features to: /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/modeling/final.csv Note: segmentation detected as NON-EMPTY; emitted observed metrics. [SUCCESS] Build canonical modeling features (schema-aligned) [STEP] Rename/copy final feature table [SUCCESS] Rename/copy final feature table ============================================================ [SECTION] Cleanup & organization (pre-cleanup staging) ============================================================ [STEP] Archive DICOM headers [SUCCESS] Archive DICOM headers [STEP] Move .dicom_header files [SUCCESS] Move .dicom_header files [STEP] Rename segmentation output → SEG.nii.gz [SUCCESS] Rename segmentation output → SEG.nii.gz [STEP] Move PET.dicom_header to preprocessed [SUCCESS] Move PET.dicom_header to preprocessed ============================================================ [SECTION] QC (pre-cleanup staging) ============================================================ [STEP] Generate QC report (staging) [SUCCESS] Generate QC report (staging) [INFO] Pipeline section finished; cleanup/report/slices will run next via trap. ----- CLEAN-UP: start (exit status: 0) ----- ============================================================ [SECTION] Collect Artifacts (clean-up) ============================================================ [INFO] Copied modeling features CSV → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/ModelingLymphoma_report/ [INFO] Placed ModelingLymphoma_SEG.nii.gz → /home/jakeaiml/Desktop/building_release/build_ingensia_pet_ai/ModelingLymphoma/ModelingLymphoma_report/ ============================================================ [SECTION] Report Generation (clean-up) ============================================================ [STEP] Build HTML report via package CLI