Breast Cancer Classification Using FCN and Beta Wavelet Autoencoder.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Biomedical Engineering

Background:

• Accurate breast cancer classification is crucial for effective treatment.
• Existing feature extraction methods for mammography analysis have limitations.
• Developing robust automated systems can improve diagnostic efficiency.

Purpose of the Study:

• To present a new breast cancer classification approach integrating Fully Convolutional Networks (FCNs) and Beta Wavelet Autoencoder (BWAE).
• To enhance feature extraction from mammography images by focusing on relevant information.
• To improve the accuracy and reliability of automated breast cancer detection.

Main Methods:

• Utilizing FCNs for image segmentation to identify relevant zones in mammography images.
• Employing BWAE for modeling extracted information, focusing on superior feature extraction capabilities.
• Combining FCNs and BWAE to refine feature extraction by retaining only pertinent features for mass identification.

Main Results:

• The proposed method demonstrated high effectiveness on a standard mammographic image dataset.
• Achieved a precision of 94% for benign and 93% for malignant cases.
• Reached a recall rate of 92% for benign and 95% for malignant cases, with 100% accuracy for normal cases.

Conclusions:

• The fusion of FCNs and BWAE significantly improves breast cancer classification accuracy.
• The proposed method shows competitive performance compared to state-of-the-art approaches.
• This technique offers a promising tool for automated analysis of mammography images.