Related Experiment Video
Updated: Jun 17, 2026
13:45
Hybrid µCT-FMT imaging and image analysis
Published on: June 4, 2015
Liver segmentation by intensity analysis and anatomical information in multi-slice CT images
Amir H Foruzan1, Reza Aghaeizadeh Zoroofi, Masatoshi Hori
1Control and Intelligent Processing Center of Excellence, School of Electrical and Computer Engineering, College of Engineering, University of Tehran, Tehran, Iran.
International Journal of Computer Assisted Radiology and Surgery
|December 25, 2009
Summary
This study presents a novel algorithm for initial liver boundary estimation in CT images. The method accurately segments the liver, aiding in diagnosis and treatment planning for liver conditions.
Area of Science:
- Medical Imaging
- Computational Anatomy
- Image Processing
Background:
- Accurate liver segmentation is crucial for clinical applications like diagnosis, treatment planning, and monitoring.
- Liver segmentation serves as a foundational step for advanced analyses such as visualization, surgical planning, and shape analysis.
Purpose of the Study:
- To propose and evaluate an algorithm for estimating an initial liver boundary from CT images.
- To establish a reliable method for liver segmentation that can be integrated into further clinical workflows.
Main Methods:
- Utilized the Expectation-Maximization (EM) algorithm to compute statistical parameters of liver intensity ranges.
- Defined a Region of Interest (ROI) by automatically extracting ribs and segmenting the heart to confine the liver.
- Employed a double thresholding approach and an anatomical rule-based system to identify and finalize liver candidates, with a color-map transformation for visual differentiation.
Main Results:
- Evaluated on local datasets, achieving an average overlap error of 15.3% and an average volume difference of 2.8%.
- Tested on MICCAI 2007 Grand Challenge datasets, yielding an average overlap error of 20.3% and an average volume difference of -4.5%.
Conclusions:
- The proposed technique demonstrates feasibility in consistently estimating initial liver borders.
- The derived liver boundary can be effectively utilized with 'Active Contour' algorithms for precise liver mask finalization.
