Registration-Based Organ Positioning and Joint Segmentation Method for Liver and Tumor Segmentation

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Biomedical Engineering

Background:

• Accurate liver and tumor segmentation in CT images is crucial for diagnosis and treatment.
• Challenges include complex anatomy and low contrast, hindering automated methods.
• Existing techniques often struggle with precision and efficiency.

Purpose of the Study:

• To develop an automated, accurate, and efficient method for segmenting liver and tumors from CT images.
• To address the limitations of current segmentation techniques.
• To improve the reliability of computer-aided diagnosis in liver disease.

Main Methods:

• A registration-based organ positioning (ROP) method for precise liver bounding box identification.
• A joint segmentation approach using fuzzy c-means (FCM) and extreme learning machine (ELM) for initial liver segmentation.
• Active contour model (ACM) refinement of liver boundaries and a separate ELM for tumor segmentation.

Main Results:

• The proposed ROP and joint segmentation method demonstrated superior performance.
• Achieved accurate liver and tumor segmentation on two experimental datasets.
• Outperformed existing related methods in segmentation accuracy and efficiency.

Conclusions:

• The developed method offers a significant advancement in automated liver and tumor segmentation from CT images.
• This technique holds promise for enhancing clinical diagnosis and treatment planning.
• The combination of ROP, FCM, ELM, and ACM provides a robust solution for complex medical image segmentation.