Suspicious-Region Segmentation From Breast Thermogram Using DLPE-Based Level Set Method.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Computer Vision

Background:

• Accurate segmentation of suspicious regions (SRs) in thermal breast images (TBIs) is crucial for breast cancer detection.
• Existing segmentation methods face challenges in precision and speed.

Purpose of the Study:

• To propose a novel active contour model for accurate and efficient segmentation of SRs in TBIs.
• To develop a framework for differentiating abnormal and normal breasts using segmented SRs.

Main Methods:

• A novel method (SCH-CS) for initial SR localization using intensity and centroid information.
• A new energy functional (DLPE) integrated into the level set method (LSM) for refined segmentation.
• Incorporation of a new stopping criterion to enhance LSM speed and accuracy.

Main Results:

• The proposed LSM achieved average segmentation accuracies of 72.18% (DMR-IR) and 71.26% (DBT-TU-JU), outperforming state-of-the-art methods.
• The novel framework effectively differentiates abnormal from normal breasts by analyzing segmented SRs.
• Analyzing only SRs proved more effective than analyzing the whole breast for differentiation.

Conclusions:

• The proposed active contour model significantly improves SR segmentation in TBIs.
• The developed framework offers a more effective approach to breast cancer screening using thermal imaging.
• Focusing on SRs enhances the diagnostic capability of TBIs.