A CNN Hyperparameters Optimization Based on Particle Swarm Optimization for Mammography Breast Cancer Classification.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Breast cancer is a leading cause of mortality in women globally.
• Medical imaging, particularly mammography, is crucial for early detection.
• Manual analysis of mammograms is time-consuming and prone to variability.

Purpose of the Study:

• To develop a highly accurate deep learning model for automated breast cancer detection in mammography.
• To address the challenge of determining optimal CNN hyperparameters and architectures.
• To enhance the efficiency and accuracy of breast cancer screening.

Main Methods:

• A novel deep learning approach utilizing Convolutional Neural Networks (CNNs).
• Integration of Particle Swarm Optimization (PSO) for automatic CNN hyperparameter and architecture selection.
• Validation on the DDSM and MIAS mammography datasets.

Main Results:

• Achieved high success rates of 98.23% on the DDSM dataset.
• Achieved high success rates of 97.98% on the MIAS dataset.
• Demonstrated superior accuracy compared to existing studies in mammography classification.

Conclusions:

• The proposed PSO-optimized CNN model offers a powerful and automated technique for breast cancer prediction.
• Automated CNN model creation for mammography classification is now feasible.
• This method significantly enhances diagnostic accuracy and efficiency in breast cancer screening.