Semi-automated enhanced breast tumor segmentation for CT image

Area of Science:

• Medical Imaging
• Radiology
• Computational Pathology

Background:

• Accurate breast cancer detection is crucial for effective treatment.
• X-ray computed tomography (CT) is a valuable tool for breast cancer diagnosis, alongside MRI and ultrasound.

Purpose of the Study:

• To propose a semi-automated breast cancer segmentation method for CT images.
• To improve the accuracy and efficiency of breast cancer region identification in medical scans.

Main Methods:

• Maximum region searching to identify initial lesion boundaries.
• Modified Histogram Equalization with Iterative-Filling for lesion enhancement and intensity balancing.
• Four-seeds Random Walk algorithm for precise segmentation of breast cancer regions.

Main Results:

• The proposed method achieved a Dice Coefficient of 88.6% on a clinical dataset of 50 cases (630 slices).
• Outperformed existing methods, with Random Walk achieving 76.9% and Graph-Cut 79.8%.

Conclusions:

• The developed semi-automated segmentation technique offers superior performance for breast cancer detection in CT images.
• This method holds potential for enhancing diagnostic accuracy and guiding treatment strategies in oncology.