Detection of cancerous masses for screening mammography using discrete wavelet transform-based multiresolution Markov...

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Biomedical Signal Processing

Background:

• Mammography is crucial for breast cancer screening.
• Dense breast tissue can obscure cancerous masses, reducing detection rates.
• Accurate detection of subtle lesions is vital for early diagnosis and treatment.

Purpose of the Study:

• To develop and evaluate a novel algorithm for enhancing the visibility of suspicious masses in mammograms.
• To improve the accuracy of computer-aided detection (CAD) systems for breast cancer.

Main Methods:

• The proposed method utilizes the discrete wavelet transform (DWT) for multi-scale image decomposition.
• A multiresolution Markov random field (MMRF) model is applied to segment and highlight suspicious regions.
• The algorithm integrates DWT and MMRF to overcome challenges posed by varying parenchymal tissue densities.

Main Results:

• The DWT-MMRF algorithm effectively isolates suspicious masses from surrounding dense tissue.
• The method demonstrates potential in improving the sensitivity of mammogram interpretation.
• Quantitative and qualitative assessments show the algorithm's capability in singling out subtle lesions.

Conclusions:

• The developed DWT-MMRF algorithm offers a promising approach to assist radiologists in detecting difficult-to-see cancerous masses.
• This technique can potentially enhance diagnostic accuracy in mammography.
• Further clinical validation is warranted to integrate this tool into routine breast cancer screening workflows.