google ads

Cinematic Rendering of Liver Diseases – Preliminary Observations

 

 

Cinematic Rendering of Liver Diseases – Preliminary Observations

Linda C. Chu

The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, MD

 

Disclosures

  • EK Fishman is the co-founder of HipGraphics and receives institutional grant support from Siemens Healthineers and GE Healthcare
  • LC Chu and SP Rowe have no disclosures

 

Learning Objectives

  • Review the algorithm of cinematic rendering and how it differs from conventional 3D volume rendering
  • Describe the potential role of cinematic rendering of liver diseases
  • Describe current limitations and future opportunities for cinematic rendering

 

Cinematic Rendering (CR)

  • Cinematic rendering (CR) is a recently described rendering technique inspired by the quality of computer animation programs
  • CR uses a global illumination model that takes direct and indirect illumination into account to generate photorealistic images
Cinematic Rendering of Liver Diseases – Preliminary Observations
Comaniciu D et al. Med Image Anal. 2016;33:19-26. Fellner FA et al. J Biomed Sci Eng. 2016;9:170-5.
Eid M et al. AJR. 2017;209(2):370-9. Johnson PT et al. AJR. 2017;209(2):309-12.

 

Volume Rendering (VR) vs. Cinematic Rendering (CR)

  • In both volume rendering (VR) and cinematic rendering (CR), thin-slice reconstructed CT data is stacked into a 3D volume
  • Each isotropic voxel in the volume is assigned a color and transparency based on attenuation thresholds
  • Then light is projected through the volume in order to generate the 3D images
Cinematic Rendering of Liver Diseases – Preliminary Observations
Calhoun PS et al. Radiographics. 1999;19(3):745-64. Fishman EK et al. Radiographics. 2006;26(3):905-22.

 

Volume Rendering (VR) vs. Cinematic Rendering (CR)

Volume Rendering (VR)
  • Simple ray casting method
  • Each pixel is formed by one light ray passing through the volumetric dataset
Cinematic Rendering (CR)
  • Monte Carlo path tracing and global illumination
  • Each pixel is formed by thousands of light rays passing through the volumetric dataset
  • Includes effects of light rays from scatter and from voxels adjacent to the paths of the rays
Comaniciu D et al. Med Image Anal. 2016;33:19-26. Fellner FA et al. J Biomed Sci Eng. 2016;9:170-5.
Eid M et al. AJR. 2017;209(2):370-9. Johnson PT et al. AJR. 2017;209(2):309-12.

 

Cinematic Rendering (CR)

  • CR generates photorealistic images that have the potential to more accurately depict anatomic detail and complex anatomy
  • A few reports have described the use of this novel technique in cardiovascular, musculoskeletal, and abdominal imaging
  • Application of this technique to liver imaging has not been previously reported in detail
Cinematic Rendering (CR)
Rowe SP et al. JCCT. 2018;12(1):56-59. Rowe SP et al. Emerg Radiol. 2018;25(1):93-101.
Chu LC et al. Abdom Radiol (NY). 2018 [Epub ahead of print]. Johnson PT et al. AJR. 2017;209(2):309-312.

 

Case Review of CR in Liver Imaging

Focal Liver Masses
  • Hemangioma
  • Focal nodular hyperplasia
  • Hepatic Adenoma
  • Abscess
  • Hepatocellular carcinoma
  • Cholangiocarcinoma
  • Metastatic Disease
Diffuse Liver Disease
  • Cirrhosis
  • Portal hypertension

 

Case 1: Hepatic Hemangioma

49 year-old woman with hepatic hemangioma.
Case 1: Hepatic Hemangioma

Axial IV contrast enhanced CT images show peripheral nodular enhancement (A) with progressive centripetal filling (B) of a classic hepatic hemangioma.

 

Case 1: Hepatic Hemangioma

49 year-old woman with hepatic hemangioma.
Case 1: Hepatic Hemangioma

Cinematic rendering (C and D) can accentuate the peripheral nodular enhancement pattern compared to 2D axial images (A and B).

 

Case 2: Giant Hepatic Hemangioma

42 year-old woman with a giant hepatic hemangioma.
Case 2: Giant Hepatic Hemangioma

Cinematic rendering shows classic peripheral nodular enhancement of a giant hepatic hemangioma (A). Translucency of the mass can be altered with the dynamic display (B and C) to appreciate this enhancement pattern.

 

Case 3: Focal Nodular Hyperplasia

52 year-old woman with no prior liver disease with incidentally diagnosed focal nodular hyperplasia.
Case 3: Focal Nodular Hyperplasia

Arterial phase coronal IV contrast enhanced CT image (A) shows a homogeneously hypervascular mass in the left hepatic lobe. Arterial phase CR image (B) shows increased conspicuity of the mass compared to 2D images. Venous phase CR image (C) shows isoenhancement of the mass relative to background liver.

 

Case 3: Focal Nodular Hyperplasia

52 year-old woman with no prior liver disease with incidentally diagnosed focal nodular hyperplasia.
Case 3: Focal Nodular Hyperplasia

Background liver can be rendered translucent to increase conspicuity of the hypervascular mass (D). CR vascular map (E) can highlight the central feeding artery and spokewheel branches, classic features of focal nodular hyperplasia. CR video (F) shows manipulation of the dynamic window display.

 

Case 4: Hepatic Adenoma

32 year-old woman with no prior liver disease with a hypervascular liver mass. Resection showed hepatic adenoma.
Case 4: Hepatic Adenoma

Axial IV contrast enhanced CT image (A) shows a heterogeneously enhancing left hepatic lobe mass in a non-cirrhotic liver. CR images can accentuate the enhancement on both arterial and venous phases (B and C) to increase tumor conspicuity.

 

Case 5: Liver Abscess

70 year-old man with sepsis and liver abscess. Fine needle aspiration returned scant fluid positive for Gram negative bacteria.
Case 5: Liver Abscess

Axial IV contrast enhanced CT image (A) shows a multiseptated cystic mass in the liver dome. CR images (B and C) can increase the appreciation of the thick enhancing internal septations, a feature that suggests the mass may be difficult to drain with percutaneous approach.

 

Case 6: Liver Abscess

49 year-old woman with liver abscess s/p pigtail placement.
Case 6: Liver Abscess

Axial IV contrast enhanced CT image (A) shows a peripherally enhancing cystic mass within the liver dome. CR images (B and C) can increase the appreciation of the thick enhancing capsule.

 

Case 7: Hepatocellular Carcinoma

79 year-old man with history of hepatitis B and C with hepatocellular carcinoma.
Case 7: Hepatocellular Carcinoma

Axial IV contrast enhanced CT images (A and B) show a subtle hypervascular mass (A) with subtle washout on venous phase (B). CR (C) can increase the conspicuity of the subtle hypervascular mass.

 

Case 8: Hepatocellular Carcinoma

84 year-old woman with biopsy proven hepatocellular carcinoma.
Case 8: Hepatocellular Carcinoma

Axial IV contrast enhanced CT image (A) shows a heterogeneous hypervascular mass in the right hepatic lobe. CR images (B) can render the background liver transparent to highlight the feeding arteries, and can potentially help in pretreatment planning for chemoembolization.

 

Case 9: Cholangiocarcinoma

75 year-old woman with biopsy proven cholangiocarcinoma.
Case 9: Cholangiocarcinoma

Coronal IV contrast enhanced CT images (A and B) show a heterogeneously enhancing liver mass (A) with progressive delayed enhancement (B). CR images (C and D) accentuate the heterogeneous internal enhancement – an important feature in distinguishing malignant from benign liver masses.

 

Case 10: Cholangiocarcinoma

60 year-old woman with history of cholangiocarcinoma.
Case 10: Cholangiocarcinoma

Axial IV contrast enhanced CT image (A) shows a heterogeneously enhancing mass with intrahepatic biliary ductal dilatation. Dynamic window display of CR images (B and C) can be adjusted to highlight the mass (B) and the dilated bile ducts (C).

 

Case 11: Mixed Cholangiocarcinoma and Hepatocellular Carcinoma

92 year-old man with biopsy-proven mixed cholangiocarcinoma and hepatocellular carcinoma with portal vein thrombosis.
Case 11: Mixed Cholangiocarcinoma and Hepatocellular Carcinoma

Coronal IV contrast enhanced CT image (A) shows a large hypoenhancing mass in the right hepatic lobe with thrombosis of portal vein branches. CR (B) can increase the conspicuity of the liver mass and portal vein thrombus.

 

Case 12: Metastatic Cholangiocarcinoma

61 year-old woman with metastatic cholangiocarcinoma.
Case 12: Metastatic Cholangiocarcinoma

Coronal IV contrast enhanced CT image (A) shows numerous enhancing liver masses with targetoid appearance. CR images (B and C) can render the background liver translucent to highlight the global disease burden.

 

Case 13: Metastatic Colon Cancer with Gallbladder Invasion

79 year-old man with history of metastatic colon cancer.
Case 13: Metastatic Colon Cancer with Gallbladder Invasion

Coronal IV contrast enhanced CT image (A) shows a liver metastasis in the right hepatic lobe with invasion of the gallbladder lumen. CR (B) can improve depth perception and increase appreciation of tumor invasion into the gallbladder.

 

Case 14: Metastatic GIST

58 year-old man with history of metastatic gastrointestinal stromal tumor.
Case 14: Metastatic GIST

Axial IV contrast enhanced CT image (A) shows a cystic mass with multiple subtle internal septations. CR (B) improves the depth perception and appreciation of subtle internal septations and internal architecture.

 

Case 15: Metastatic Head and Neck Squamous Cell Carcinoma

45 year-old man with metastatic head and neck squamous cell carcinoma.
Case 15: Metastatic Head and Neck Squamous Cell Carcinoma

Coronal IV contrast enhanced CT image (A) shows numerous hypoenhancing liver lesions compatible with metastatic disease. CR image (B) can increase tissue contrast and increase the conspicuity of these metastatic lesions.

 

Potential Benefits of CR in Evaluation of Focal Liver Masses

  • Improve detection of subtle mass lesion
  • Improve appreciation of enhancement pattern
    • Can narrow the differential diagnosis
  • Improve appreciation of internal architecture
    • Can differentiate between benign and aggressive lesions

Potential Benefits of CR in Evaluation of Focal Liver Masses

 

Potential Benefits of CR in Evaluation of Focal Liver Masses

  • Improve appreciation of extent of local tumor extension
    • Can assist in preoperative planning
  • Improve appreciation of feeding vessels
    • Can assist in chemoembolization planning
  • Improve appreciation of global disease burden
    • Facilitate communication with referring clinicians and patients

Potential Benefits of CR in Evaluation of Focal Liver Masses

 

Case 16: Cirrhosis

61 year-old man with history of cirrhosis.
Case 16: Cirrhosis

Axial IV contrast enhanced CT image (A) shows diffuse nodular parenchymal appearance of a cirrhotic liver. CR images (B and C) show greater appreciation of the nodular parenchymal texture due to improved depth perception and photorealism with CR.

 

Case 17: Cirrhosis with Portal Hypertension

64 year-old woman with history of cirrhosis and portal hypertension.
Case 17: Cirrhosis with Portal Hypertension

CR allows the user to alter the window settings to highlight the anatomy of interest. In this case, the extensive portal systemic collaterals within the left lower quadrant can be highlighted in the vascular map (B) by rendering the overlapping organs transparent.

 

Case 18: Cirrhosis with Portal Hypertension

69 year-old woman with history of cirrhosis s/p liver transplant with mild intrahepatic biliary dilatation, splenomegaly, and varices.
Case 18: Cirrhosis with Portal Hypertension

Axial IV contrast enhanced CT image (A) shows marked left upper quadrant varices from portal hypertension. CR image (B) can show the varices and splenomegaly with increased photorealistic detail.

 

Discussion

  • This case review shows the potential application of cinematic rendering (CR) in the evaluation of focal liver masses and diffuse parenchymal disease
  • CR has the potential to increase lesion conspicuity, accentuate enhancement pattern, and assist in pretreatment planning
  • CR can illustrate cirrhosis and sequel of portal hypertension with increased photorealistic detail

 

Future Directions

  • These results will require further validation in larger studies
  • Future directions:
    • Evaluate the role of photorealism and improved anatomic detail in detection and classification of focal liver masses
    • Evaluate whether CR vascular maps assist in preoperative planning in chemoembolization of liver masses

 

Limitations

  • CR uses a more complex algorithm than VR and requires higher computational power
    • Requires more powerful workstations and network bandwidth
    • Each manipulation during interactive rendering requires a new set of calculations → Longer post-processing time compared to VR
  • Needs to be performed in real-time at the current time and should be done by the radiologist
  • Further research is needed to determine whether the photorealism improves diagnostic accuracy and/or outcomes

 

Conclusion

  • Cinematic rendering is a recently described post-processing technique that generates photorealistic images from CT datasets
  • It has the potential to more accurately depict anatomic detail and complex anatomy
  • The preliminary application in liver imaging shows potential benefits of CR in the evaluation of focal liver masses and diffuse parenchymal disease
  • These results will require validation in larger studies

 

References

  • Calhoun PS et al. Radiographics. 1999;19(3):745-64.
  • Chu LC et al. Abdom Radiol (NY). 2018 [Epub ahead of print].
  • Comaniciu D et al. Med Image Anal. 2016;33:19-26.
  • Eid M et al. AJR. 2017;209(2):370-9.
  • Fellner FA et al. J Biomed Sci Eng. 2016;9:170-5.
  • Fishman EK et al. Radiographics. 2006;26(3):905-22.
  • Johnson PT et al. AJR. 2017;209(2):309-12.
  • Rowe SP et al. JCCT. 2018;12(1):56-59.
  • Rowe SP et al. Emerg Radiol. 2018;25(1):93-101.

Acknowledgements

  • Linda C. Chu
  • Steven P. Rowe
  • Elliot K. Fishman

Privacy Policy

Copyright © 2024 The Johns Hopkins University, The Johns Hopkins Hospital, and The Johns Hopkins Health System Corporation. All rights reserved.