Suspicious Lesion Segmentation on Brain, Mammograms and Breast MR Images Using New Optimized Spatial Feature Based...

Area of Science:

• Medical Image Analysis
• Computer-Aided Diagnosis
• Artificial Intelligence in Medicine

Background:

• Suspicious lesion and organ segmentation is critical for medical diagnosis but faces challenges like low accuracy and robustness.
• Traditional Fuzzy C-Means Clustering (FCM) is sensitive to noise due to its neglect of spatial information.

Purpose of the Study:

• To develop a more accurate and robust medical image segmentation method.
• To address the limitations of existing segmentation techniques in handling noise and complexity.

Main Methods:

• Implementation of a super-pixel-based Fuzzy C-Means Clustering (SPOFCM) algorithm.
• Incorporation of spatial information from neighboring super-pixels.
• Optimization of the influential degree using a crow search algorithm.

Main Results:

• The proposed SPOFCM method demonstrated improved segmentation performance.
• Feasibility verified on multi-spectral MRIs and mammograms for tumor segmentation.
• Outperformed k-means, entropy thresholding, FCM, FCM_S, and KFCM in comparative tests.

Conclusions:

• SPOFCM offers enhanced accuracy and robustness for suspicious lesion and organ segmentation.
• The algorithm shows significant potential for computer-assisted clinical applications.
• This method improves upon traditional FCM by effectively utilizing spatial information.