Computer-aided diagnosis and quantification of cirrhotic livers based on morphological analysis and machine learning

Area of Science:

• Medical Imaging
• Computational Anatomy
• Hepatology

Background:

• Morphological changes in the liver and spleen are characteristic of chronic liver diseases.
• Accurate quantification and staging of these changes are crucial for patient management.

Purpose of the Study:

• To develop a computer-aided diagnosis (CAD) method for chronic liver disease using statistical shape models (SSMs).
• To analyze liver and spleen morphology in CT images for improved diagnosis and staging.

Main Methods:

• Construction of individual liver and spleen SSMs, as well as a joint SSM.
• Selection of effective morphological modes based on contribution rate and correlation with clinical staging.
• Learning a mapping function between selected modes and chronic liver disease stages.

Main Results:

• The developed mapping function demonstrated utility in diagnosing and staging chronic liver diseases.
• The SSM approach provides a quantitative method for assessing morphological alterations.

Conclusions:

• Statistical shape models offer a promising approach for computer-aided analysis of chronic liver disease.
• This method can enhance the accuracy of diagnosis and staging of liver conditions based on morphological changes.