An iterative Bayesian approach for liver segmentation: algorithm and
clinical validation study

Purpose
We present a new algorithm for nearly automatic liver segmentation and volume estimation from abdominal Computed Tomography Angiography (CTA) images and its validation.

Materials and method
Our hybrid algorithm uses a multiresolution iterative scheme. It starts from a single user-defined pixel seed inside the liver, and repeatedly applies smoothed Bayesian classification to identify the liver and the other organs, followed by adaptive morphological operations and active contours refinement.We evaluate the algorithm with two retrospective studies on 56 validated CTA images. The first study compares it to ground-truth manual segmentation and to semi-automatic and automatic commercial methods. The second study uses the public dataset SLIVER07 and its comparison methodology.

Results
We achieved for both studies correlations of 0.98 and 0.99 for liver volume estimation, with mean volume differences of 5.36% and 2.68% with respect to manual ground-truth estimation, and mean volume variability for different initial seeds of 0.54% and 0.004%, respectively. For the second study, our algorithm scored 71.8 and 67.87 for the training and test datasets, which compares very favorably with other semi-automatic methods.

Conclusions
Our algorithm requires minimal interaction by a non-expert user, is accurate, efficient, and robust to initial seed selection. It can be effective for hepatic volume estimation and liver modeling in a clinical setup.

• Freiman. M. Eliassaf. O. Taieb. Y. Joskowicz. L. Azraq. Y. and Sosna. J. 
  "An iterative Bayesian approach to liver segmentation: algorithm and validation"
  Int. Journal of Computer Assisted Radiology and Surgery  (pdf).
  
  
• Freiman. M. Eliassaf. O. Taieb. Y. Joskowicz. L. and Sosna. J.
  "A Bayesian approach for liver analysis: algorithm and validation study",
  In Proc. of the 11th Int. Conf. of Medical Image Computing and Computed Aided Interventions, MICCAI'08, 2008 (pdf).
  
  
• Eliassaf. O. Taieb. Y. Freiman. M. Joskowicz. L. and Sosna. J.
  "An iterative Bayesian approach for liver analysis: tumors validation study",
  In Proc. of the MICCAI 3D Segmentation in the Clinic: A Grand Challenge II workshop, in MICCAI'08, 2008 (pdf). 

• Freiman. M. Eliassaf. O. Taieb. Y. Joskowicz. L. Azraq. Y. and Sosna. J.
  "An iterative Bayesian approach to liver segmentation: algorithm and clinical validation study",
  In proc. of the 22nd Int. Conf. on Computer Aided Radiology and Surgery, CARS'08, 2008.  (abstract)

Related materials:

• ISRACAS'08 presentation (ppt)
• View movies:
  • 2D axial view of the segmented slices: mpeg.
  • 3D view of the segmented liver:  mpeg.
  • 2D axial view of the segmented slices (Liver only): avi.
  • 3D view of the segmented liver (Liver only):  avi.

Links: