Computer aided system for segmentation and visualization of microcalcifications in digital mammograms

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Digital Mammography

Background:

• Accurate segmentation and visualization of microcalcifications are crucial for early breast cancer detection.
• Existing methods may face challenges with background tissue variations and contrast levels.

Purpose of the Study:

• To develop and evaluate two novel methods for microcalcification segmentation and visualization in mammograms.
• To integrate these methods into a user-friendly computer-aided visualization system for radiologists.

Main Methods:

• Mathematical Morphology: Utilizes morphological operations for high contrast enhancement and background suppression, emphasizing small bright details indicative of microcalcifications.
• Multifractal Approach: Generates multifractal images from mammograms, allowing radiologists to adjust segmentation levels interactively.
• Computer-Aided Visualization (CAV): An interactive system embedding both methods, enabling physicians to control segmentation quality and level.

Main Results:

• Both mathematical morphology and multifractal approaches demonstrated effectiveness in segmenting potential microcalcifications.
• The CAV system provided interactive control, enhancing the physician's ability to assess segmentation results.
• Methods were validated using mammograms from the MIAS database and clinical practice.

Conclusions:

• The described methods offer robust tools for microcalcification segmentation and visualization in digital mammography.
• The interactive CAV system supports radiologists in improving diagnostic accuracy for breast cancer screening.
• These advancements contribute to more effective computer-aided diagnosis in mammography.